cloudflare
tcp/443 tcp/80 tcp/8443
nginx
tcp/443
web
tcp/443 tcp/80
Open service 2a06:98c1:3121::3:443 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 200 OK
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Access-Control-Allow-Origin: *
Cache-Control: public, max-age=0, must-revalidate
cf-cache-status: DYNAMIC
Link: <https://fonts.googleapis.com>; rel="preconnect", <https://maps.googleapis.com>; rel="preconnect"
referrer-policy: strict-origin-when-cross-origin
x-content-type-options: nosniff
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=P6KoISwwMuCkGyjFqVIZGKusEXKQtxZP8MuMY7cq2f%2F%2BhiH%2FCsLux%2FRUqOV86yCS2guX2cy35SkclCtYX1NNnyTVlWWXNwGCZQz9s066PgDdgOhKanZXkmGYT0dBmSBC9%2Fc%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
Server-Timing: cfCacheStatus;desc="DYNAMIC"
Server-Timing: cfEdge;dur=8,cfOrigin;dur=71
CF-RAY: 9bb70807bf0da1af-SIN
alt-svc: h3=":443"; ma=86400
Page title: Yin Jun & Reina
<!doctype html>
<html>
<!-- Handcrafted with love -->
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1">
<link href="/output.css" rel="stylesheet">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://maps.googleapis.com">
<link href="https://fonts.googleapis.com/css2?family=Explora&family=Felipa&display=swap" rel="stylesheet">
<script async src="https://maps.googleapis.com/maps/api/js?key=AIzaSyA5PST2icMw-LMcr7CueOg8IEDe_-7ZJjU&callback=console.debug&libraries=maps,marker&v=beta"></script>
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<title>Yin Jun & Reina</title>
<script type="text/javascript" src="/invitation.js"></script>
</head>
<body class="relative">
<div class="bg-white">
<div class="relative isolate px-6 lg:pt-14 lg:px-8">
<div class="absolute inset-x-0 -z-10 transform-gpu blur-sm overflow-hidden min-[320px]:-top-20 md:-top-40 sm:-top-20" aria-hidden="true">
<div class="relative w-screen">
<picture>
<source srcset="/IMG_6646.webp" type="image/webp">
<img src="/IMG_6646.png" />
</picture>
</div>
</div>
<div class="mx-auto max-w-2xl sm:py-16 py-8 lg:py-56 md:py-32">
<div class="text-center">
<h1 class="mt-4 text-4xl font-bold tracking-tight text-slate-100 sm:text-6xl felipa-regular" style="text-shadow: 1px 1px 2px black">Yin Jun & Reina</h1>
<p class="mt-16 text-lg leading-8 text-slate-100 font-serif" style="text-shadow: 1px 1px 2px black">素敵な一日</p>
</div>
</div>
</div>
<div class="flex flex-col">
<section>
<div class="bg-white relative overflow-hidden">
<div class="border-4 border-orange-200 m-8 py-8 md:py-32">
<div class="mx-auto py-8 text-center">
<h2 class="text-md text-3xl felipa-regular">Message</h2>
</div>
<div class="mx-auto py-8">
<p class="text-center md:hidden">English follows Japanese.</p>
<div class="mx-auto p-8 md:p-14 max-w-2xl font-serif md:flex">
<div class="md:w-2/5 w-full">
<p class="mb-8">拝啓</p>
<p>皆様いかがお過ごしでしょうか</p>
<p>このたび結婚式を</p>
<p>執り行うこととなりました</p>
<p class="mt-4">日頃お世話になっております皆様に</p>
<p>私どもの門出を</p>
<p>お見守りいただきたく</p>
<p>ささやかながら小宴を</p>
<p>催したく存じます</p>
<p class="mt-4">ご多用中誠に恐縮ではございますが</p>
<p>ぜひご出席いただきたく</p>
<p>ご案内申し上げます</p>
<p class="my-8">敬具</p>
<p>ポアインジュン・椿原怜奈</p>
</div>
<div class="md:w-2/5 md:mt-0 mt-12 w-full md:ml-16 lg:ml-16 italic">
<p>It is with immense pleasure and gratitude that</p>
<p>Phua Yin Jun</p>
<p>and</p>
<p>Tsubakihara Reina</p>
<p>would like to invite you</p>
<p>to share a special moment of our lives.</p>
<p class="mt-4">We greatly look forward</p>
<p>to having the honor of your presence</p>
<p>as we write our own love story</p>
<p class="mt-4">We hope that</p>
<p>you will join us</p>
<p>and make our joy complete</p>
Open service 188.114.97.3:80 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 301 Moved Permanently
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Length: 0
Connection: close
Location: https://wedding.yinjunphua.com/
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=PVVIoW0PSFP%2BshJ2SlGmwdHB8SoO1%2BgiymPQlMsZEiC1NqU1RviVPPvzkfEi0sOMpakAxlbTt9ZZumQLqLd7PJojv9vkWf7IFvy%2B5iKikqep5jklKug%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
Server-Timing: cfCacheStatus;desc="DYNAMIC"
Server-Timing: cfEdge;dur=13,cfOrigin;dur=7
cf-cache-status: DYNAMIC
CF-RAY: 9bb70804db0bf8e0-SIN
alt-svc: h3=":443"; ma=86400
Open service 2a06:98c1:3120::3:80 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 301 Moved Permanently
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Length: 0
Connection: close
Location: https://wedding.yinjunphua.com/
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=DgDG9IHxKNqlbBPw4iy4VtAt%2FFCo8rhypYpdzczn5ZLdQMkXHcS2r3pBz%2F%2Fp%2F864oSpr5eFrsxsuWd8y%2F7a%2BheWOh5EpXyda2ieO3PiQvBfwu1ycIJE5AKnULJasMngRyrU%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
Server-Timing: cfCacheStatus;desc="DYNAMIC"
Server-Timing: cfEdge;dur=26,cfOrigin;dur=10
cf-cache-status: DYNAMIC
CF-RAY: 9bb708052cf58c81-EWR
alt-svc: h3=":443"; ma=86400
Open service 2a06:98c1:3121::3:8443 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 200 OK
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Access-Control-Allow-Origin: *
Cache-Control: public, max-age=0, must-revalidate
Link: <https://fonts.googleapis.com>; rel="preconnect", <https://maps.googleapis.com>; rel="preconnect"
referrer-policy: strict-origin-when-cross-origin
x-content-type-options: nosniff
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=Gc%2B0soiu%2F5BFET1KlI8yZG3ah3dvJribC7CKAxta6xlGiekwPwYIW2xRHZtA%2FsLBY5Gh5s5iBTorUfhBjb8IoAJDY1rUJnsPpNZFTHePpsch60YCP4%2BrpOH%2F2MDjFt1a%2FpY%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
cf-cache-status: DYNAMIC
CF-RAY: 9bb70804d8d588ce-AMS
alt-svc: h3=":8443"; ma=86400
Page title: Yin Jun & Reina
<!doctype html>
<html>
<!-- Handcrafted with love -->
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1">
<link href="/output.css" rel="stylesheet">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://maps.googleapis.com">
<link href="https://fonts.googleapis.com/css2?family=Explora&family=Felipa&display=swap" rel="stylesheet">
<script async src="https://maps.googleapis.com/maps/api/js?key=AIzaSyA5PST2icMw-LMcr7CueOg8IEDe_-7ZJjU&callback=console.debug&libraries=maps,marker&v=beta"></script>
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<title>Yin Jun & Reina</title>
<script type="text/javascript" src="/invitation.js"></script>
</head>
<body class="relative">
<div class="bg-white">
<div class="relative isolate px-6 lg:pt-14 lg:px-8">
<div class="absolute inset-x-0 -z-10 transform-gpu blur-sm overflow-hidden min-[320px]:-top-20 md:-top-40 sm:-top-20" aria-hidden="true">
<div class="relative w-screen">
<picture>
<source srcset="/IMG_6646.webp" type="image/webp">
<img src="/IMG_6646.png" />
</picture>
</div>
</div>
<div class="mx-auto max-w-2xl sm:py-16 py-8 lg:py-56 md:py-32">
<div class="text-center">
<h1 class="mt-4 text-4xl font-bold tracking-tight text-slate-100 sm:text-6xl felipa-regular" style="text-shadow: 1px 1px 2px black">Yin Jun & Reina</h1>
<p class="mt-16 text-lg leading-8 text-slate-100 font-serif" style="text-shadow: 1px 1px 2px black">素敵な一日</p>
</div>
</div>
</div>
<div class="flex flex-col">
<section>
<div class="bg-white relative overflow-hidden">
<div class="border-4 border-orange-200 m-8 py-8 md:py-32">
<div class="mx-auto py-8 text-center">
<h2 class="text-md text-3xl felipa-regular">Message</h2>
</div>
<div class="mx-auto py-8">
<p class="text-center md:hidden">English follows Japanese.</p>
<div class="mx-auto p-8 md:p-14 max-w-2xl font-serif md:flex">
<div class="md:w-2/5 w-full">
<p class="mb-8">拝啓</p>
<p>皆様いかがお過ごしでしょうか</p>
<p>このたび結婚式を</p>
<p>執り行うこととなりました</p>
<p class="mt-4">日頃お世話になっております皆様に</p>
<p>私どもの門出を</p>
<p>お見守りいただきたく</p>
<p>ささやかながら小宴を</p>
<p>催したく存じます</p>
<p class="mt-4">ご多用中誠に恐縮ではございますが</p>
<p>ぜひご出席いただきたく</p>
<p>ご案内申し上げます</p>
<p class="my-8">敬具</p>
<p>ポアインジュン・椿原怜奈</p>
</div>
<div class="md:w-2/5 md:mt-0 mt-12 w-full md:ml-16 lg:ml-16 italic">
<p>It is with immense pleasure and gratitude that</p>
<p>Phua Yin Jun</p>
<p>and</p>
<p>Tsubakihara Reina</p>
<p>would like to invite you</p>
<p>to share a special moment of our lives.</p>
<p class="mt-4">We greatly look forward</p>
<p>to having the honor of your presence</p>
<p>as we write our own love story</p>
<p class="mt-4">We hope that</p>
<p>you will join us</p>
<p>and make our joy complete</p>
Open service 2a06:98c1:3120::3:443 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 200 OK
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Access-Control-Allow-Origin: *
Cache-Control: public, max-age=0, must-revalidate
cf-cache-status: DYNAMIC
Link: <https://fonts.googleapis.com>; rel="preconnect", <https://maps.googleapis.com>; rel="preconnect"
referrer-policy: strict-origin-when-cross-origin
x-content-type-options: nosniff
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=uk22PvR4%2FAI3S2UWqmAniQvl4wjBGgVmfrPyb64N6N8vBWi5o9Td%2B%2FbJUuorPzzIfDNSKjlJf8ificc5pmIIum6w58jhB%2Bj7bsBG0QsTcPr85eI5nlrYF%2Ba%2Fc2sEXgJhGiI%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
Server-Timing: cfCacheStatus;desc="DYNAMIC"
Server-Timing: cfEdge;dur=10,cfOrigin;dur=36
CF-RAY: 9bb70804b951f795-EWR
alt-svc: h3=":443"; ma=86400
Page title: Yin Jun & Reina
<!doctype html>
<html>
<!-- Handcrafted with love -->
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1">
<link href="/output.css" rel="stylesheet">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://maps.googleapis.com">
<link href="https://fonts.googleapis.com/css2?family=Explora&family=Felipa&display=swap" rel="stylesheet">
<script async src="https://maps.googleapis.com/maps/api/js?key=AIzaSyA5PST2icMw-LMcr7CueOg8IEDe_-7ZJjU&callback=console.debug&libraries=maps,marker&v=beta"></script>
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<title>Yin Jun & Reina</title>
<script type="text/javascript" src="/invitation.js"></script>
</head>
<body class="relative">
<div class="bg-white">
<div class="relative isolate px-6 lg:pt-14 lg:px-8">
<div class="absolute inset-x-0 -z-10 transform-gpu blur-sm overflow-hidden min-[320px]:-top-20 md:-top-40 sm:-top-20" aria-hidden="true">
<div class="relative w-screen">
<picture>
<source srcset="/IMG_6646.webp" type="image/webp">
<img src="/IMG_6646.png" />
</picture>
</div>
</div>
<div class="mx-auto max-w-2xl sm:py-16 py-8 lg:py-56 md:py-32">
<div class="text-center">
<h1 class="mt-4 text-4xl font-bold tracking-tight text-slate-100 sm:text-6xl felipa-regular" style="text-shadow: 1px 1px 2px black">Yin Jun & Reina</h1>
<p class="mt-16 text-lg leading-8 text-slate-100 font-serif" style="text-shadow: 1px 1px 2px black">素敵な一日</p>
</div>
</div>
</div>
<div class="flex flex-col">
<section>
<div class="bg-white relative overflow-hidden">
<div class="border-4 border-orange-200 m-8 py-8 md:py-32">
<div class="mx-auto py-8 text-center">
<h2 class="text-md text-3xl felipa-regular">Message</h2>
</div>
<div class="mx-auto py-8">
<p class="text-center md:hidden">English follows Japanese.</p>
<div class="mx-auto p-8 md:p-14 max-w-2xl font-serif md:flex">
<div class="md:w-2/5 w-full">
<p class="mb-8">拝啓</p>
<p>皆様いかがお過ごしでしょうか</p>
<p>このたび結婚式を</p>
<p>執り行うこととなりました</p>
<p class="mt-4">日頃お世話になっております皆様に</p>
<p>私どもの門出を</p>
<p>お見守りいただきたく</p>
<p>ささやかながら小宴を</p>
<p>催したく存じます</p>
<p class="mt-4">ご多用中誠に恐縮ではございますが</p>
<p>ぜひご出席いただきたく</p>
<p>ご案内申し上げます</p>
<p class="my-8">敬具</p>
<p>ポアインジュン・椿原怜奈</p>
</div>
<div class="md:w-2/5 md:mt-0 mt-12 w-full md:ml-16 lg:ml-16 italic">
<p>It is with immense pleasure and gratitude that</p>
<p>Phua Yin Jun</p>
<p>and</p>
<p>Tsubakihara Reina</p>
<p>would like to invite you</p>
<p>to share a special moment of our lives.</p>
<p class="mt-4">We greatly look forward</p>
<p>to having the honor of your presence</p>
<p>as we write our own love story</p>
<p class="mt-4">We hope that</p>
<p>you will join us</p>
<p>and make our joy complete</p>
Open service 2a06:98c1:3121::3:80 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 301 Moved Permanently
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Length: 0
Connection: close
Location: https://wedding.yinjunphua.com/
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=Nn58EtYqd5B9YXwA4bwS%2BFa9p0dDfH32QhWda7cZ6gdhXjfcXIsDV7hhYxNiF1Owuq8OUoVQ%2FA8%2BdLj7yRhn65MYoWuzMiqlMJ3od%2Fe0v%2BquxzwO4W%2BiUgBKsN3MjiUO4dI%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
cf-cache-status: DYNAMIC
CF-RAY: 9bb7080489c3d396-FRA
alt-svc: h3=":443"; ma=86400
Open service 188.114.97.3:8443 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 200 OK
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Access-Control-Allow-Origin: *
Cache-Control: public, max-age=0, must-revalidate
cf-cache-status: DYNAMIC
Link: <https://fonts.googleapis.com>; rel="preconnect", <https://maps.googleapis.com>; rel="preconnect"
referrer-policy: strict-origin-when-cross-origin
x-content-type-options: nosniff
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=MPEG2AV4Jq9lLdis2ukgL1UoIMqISq4mco8sN0U4Lk%2Fpi5oD3ZfWW1wfa0kl%2BJZM%2BiLPW2Y0EXgcaqylwGpD7m70kDCVzavA8xmFMFbT2tCCMiW8"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
Server-Timing: cfCacheStatus;desc="DYNAMIC"
Server-Timing: cfEdge;dur=171,cfOrigin;dur=76
CF-RAY: 9bb70806a8f05fbb-SIN
alt-svc: h3=":8443"; ma=86400
Page title: Yin Jun & Reina
<!doctype html>
<html>
<!-- Handcrafted with love -->
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1">
<link href="/output.css" rel="stylesheet">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://maps.googleapis.com">
<link href="https://fonts.googleapis.com/css2?family=Explora&family=Felipa&display=swap" rel="stylesheet">
<script async src="https://maps.googleapis.com/maps/api/js?key=AIzaSyA5PST2icMw-LMcr7CueOg8IEDe_-7ZJjU&callback=console.debug&libraries=maps,marker&v=beta"></script>
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<title>Yin Jun & Reina</title>
<script type="text/javascript" src="/invitation.js"></script>
</head>
<body class="relative">
<div class="bg-white">
<div class="relative isolate px-6 lg:pt-14 lg:px-8">
<div class="absolute inset-x-0 -z-10 transform-gpu blur-sm overflow-hidden min-[320px]:-top-20 md:-top-40 sm:-top-20" aria-hidden="true">
<div class="relative w-screen">
<picture>
<source srcset="/IMG_6646.webp" type="image/webp">
<img src="/IMG_6646.png" />
</picture>
</div>
</div>
<div class="mx-auto max-w-2xl sm:py-16 py-8 lg:py-56 md:py-32">
<div class="text-center">
<h1 class="mt-4 text-4xl font-bold tracking-tight text-slate-100 sm:text-6xl felipa-regular" style="text-shadow: 1px 1px 2px black">Yin Jun & Reina</h1>
<p class="mt-16 text-lg leading-8 text-slate-100 font-serif" style="text-shadow: 1px 1px 2px black">素敵な一日</p>
</div>
</div>
</div>
<div class="flex flex-col">
<section>
<div class="bg-white relative overflow-hidden">
<div class="border-4 border-orange-200 m-8 py-8 md:py-32">
<div class="mx-auto py-8 text-center">
<h2 class="text-md text-3xl felipa-regular">Message</h2>
</div>
<div class="mx-auto py-8">
<p class="text-center md:hidden">English follows Japanese.</p>
<div class="mx-auto p-8 md:p-14 max-w-2xl font-serif md:flex">
<div class="md:w-2/5 w-full">
<p class="mb-8">拝啓</p>
<p>皆様いかがお過ごしでしょうか</p>
<p>このたび結婚式を</p>
<p>執り行うこととなりました</p>
<p class="mt-4">日頃お世話になっております皆様に</p>
<p>私どもの門出を</p>
<p>お見守りいただきたく</p>
<p>ささやかながら小宴を</p>
<p>催したく存じます</p>
<p class="mt-4">ご多用中誠に恐縮ではございますが</p>
<p>ぜひご出席いただきたく</p>
<p>ご案内申し上げます</p>
<p class="my-8">敬具</p>
<p>ポアインジュン・椿原怜奈</p>
</div>
<div class="md:w-2/5 md:mt-0 mt-12 w-full md:ml-16 lg:ml-16 italic">
<p>It is with immense pleasure and gratitude that</p>
<p>Phua Yin Jun</p>
<p>and</p>
<p>Tsubakihara Reina</p>
<p>would like to invite you</p>
<p>to share a special moment of our lives.</p>
<p class="mt-4">We greatly look forward</p>
<p>to having the honor of your presence</p>
<p>as we write our own love story</p>
<p class="mt-4">We hope that</p>
<p>you will join us</p>
<p>and make our joy complete</p>
Open service 188.114.97.3:443 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 200 OK
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Access-Control-Allow-Origin: *
Cache-Control: public, max-age=0, must-revalidate
cf-cache-status: DYNAMIC
Link: <https://fonts.googleapis.com>; rel="preconnect", <https://maps.googleapis.com>; rel="preconnect"
referrer-policy: strict-origin-when-cross-origin
x-content-type-options: nosniff
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=oRJuIahI3z6tEYFAYYi9t4z9%2FGOlsHeEiwujvWi4My531zav3zy%2BTHHcT9jbRkqrIiM963LwgzsheM0Pd81qRW7JRKU4Ve%2BLJLEKOkVT1Xw2zruyouc%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
Server-Timing: cfCacheStatus;desc="DYNAMIC"
Server-Timing: cfEdge;dur=163,cfOrigin;dur=36
CF-RAY: 9bb70804bfd9f52e-EWR
alt-svc: h3=":443"; ma=86400
Page title: Yin Jun & Reina
<!doctype html>
<html>
<!-- Handcrafted with love -->
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1">
<link href="/output.css" rel="stylesheet">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://maps.googleapis.com">
<link href="https://fonts.googleapis.com/css2?family=Explora&family=Felipa&display=swap" rel="stylesheet">
<script async src="https://maps.googleapis.com/maps/api/js?key=AIzaSyA5PST2icMw-LMcr7CueOg8IEDe_-7ZJjU&callback=console.debug&libraries=maps,marker&v=beta"></script>
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<title>Yin Jun & Reina</title>
<script type="text/javascript" src="/invitation.js"></script>
</head>
<body class="relative">
<div class="bg-white">
<div class="relative isolate px-6 lg:pt-14 lg:px-8">
<div class="absolute inset-x-0 -z-10 transform-gpu blur-sm overflow-hidden min-[320px]:-top-20 md:-top-40 sm:-top-20" aria-hidden="true">
<div class="relative w-screen">
<picture>
<source srcset="/IMG_6646.webp" type="image/webp">
<img src="/IMG_6646.png" />
</picture>
</div>
</div>
<div class="mx-auto max-w-2xl sm:py-16 py-8 lg:py-56 md:py-32">
<div class="text-center">
<h1 class="mt-4 text-4xl font-bold tracking-tight text-slate-100 sm:text-6xl felipa-regular" style="text-shadow: 1px 1px 2px black">Yin Jun & Reina</h1>
<p class="mt-16 text-lg leading-8 text-slate-100 font-serif" style="text-shadow: 1px 1px 2px black">素敵な一日</p>
</div>
</div>
</div>
<div class="flex flex-col">
<section>
<div class="bg-white relative overflow-hidden">
<div class="border-4 border-orange-200 m-8 py-8 md:py-32">
<div class="mx-auto py-8 text-center">
<h2 class="text-md text-3xl felipa-regular">Message</h2>
</div>
<div class="mx-auto py-8">
<p class="text-center md:hidden">English follows Japanese.</p>
<div class="mx-auto p-8 md:p-14 max-w-2xl font-serif md:flex">
<div class="md:w-2/5 w-full">
<p class="mb-8">拝啓</p>
<p>皆様いかがお過ごしでしょうか</p>
<p>このたび結婚式を</p>
<p>執り行うこととなりました</p>
<p class="mt-4">日頃お世話になっております皆様に</p>
<p>私どもの門出を</p>
<p>お見守りいただきたく</p>
<p>ささやかながら小宴を</p>
<p>催したく存じます</p>
<p class="mt-4">ご多用中誠に恐縮ではございますが</p>
<p>ぜひご出席いただきたく</p>
<p>ご案内申し上げます</p>
<p class="my-8">敬具</p>
<p>ポアインジュン・椿原怜奈</p>
</div>
<div class="md:w-2/5 md:mt-0 mt-12 w-full md:ml-16 lg:ml-16 italic">
<p>It is with immense pleasure and gratitude that</p>
<p>Phua Yin Jun</p>
<p>and</p>
<p>Tsubakihara Reina</p>
<p>would like to invite you</p>
<p>to share a special moment of our lives.</p>
<p class="mt-4">We greatly look forward</p>
<p>to having the honor of your presence</p>
<p>as we write our own love story</p>
<p class="mt-4">We hope that</p>
<p>you will join us</p>
<p>and make our joy complete</p>
Open service 2a06:98c1:3120::3:8443 · wedding.yinjunphua.com
2026-01-09 21:29
HTTP/1.1 200 OK
Date: Fri, 09 Jan 2026 21:29:39 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Access-Control-Allow-Origin: *
Cache-Control: public, max-age=0, must-revalidate
Link: <https://fonts.googleapis.com>; rel="preconnect", <https://maps.googleapis.com>; rel="preconnect"
referrer-policy: strict-origin-when-cross-origin
x-content-type-options: nosniff
Vary: accept-encoding
Report-To: {"group":"cf-nel","max_age":604800,"endpoints":[{"url":"https://a.nel.cloudflare.com/report/v4?s=DLMX9l6AH7n3lo%2BIusxQewQ1Pb%2BSjDI2jEXTkZjSUy9etwd%2Ben%2Bgf4gSQJJURQeKaRiAgp8tPyIAeTwhFUP8NlCkyly6NoMi0fThG8Vbe9%2Bb2cU5v4U2Ch3GCSt982CbdfY%3D"}]}
Nel: {"report_to":"cf-nel","success_fraction":0.0,"max_age":604800}
Server: cloudflare
cf-cache-status: DYNAMIC
CF-RAY: 9bb70804d8b21963-FRA
alt-svc: h3=":8443"; ma=86400
Page title: Yin Jun & Reina
<!doctype html>
<html>
<!-- Handcrafted with love -->
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1">
<link href="/output.css" rel="stylesheet">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://maps.googleapis.com">
<link href="https://fonts.googleapis.com/css2?family=Explora&family=Felipa&display=swap" rel="stylesheet">
<script async src="https://maps.googleapis.com/maps/api/js?key=AIzaSyA5PST2icMw-LMcr7CueOg8IEDe_-7ZJjU&callback=console.debug&libraries=maps,marker&v=beta"></script>
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<title>Yin Jun & Reina</title>
<script type="text/javascript" src="/invitation.js"></script>
</head>
<body class="relative">
<div class="bg-white">
<div class="relative isolate px-6 lg:pt-14 lg:px-8">
<div class="absolute inset-x-0 -z-10 transform-gpu blur-sm overflow-hidden min-[320px]:-top-20 md:-top-40 sm:-top-20" aria-hidden="true">
<div class="relative w-screen">
<picture>
<source srcset="/IMG_6646.webp" type="image/webp">
<img src="/IMG_6646.png" />
</picture>
</div>
</div>
<div class="mx-auto max-w-2xl sm:py-16 py-8 lg:py-56 md:py-32">
<div class="text-center">
<h1 class="mt-4 text-4xl font-bold tracking-tight text-slate-100 sm:text-6xl felipa-regular" style="text-shadow: 1px 1px 2px black">Yin Jun & Reina</h1>
<p class="mt-16 text-lg leading-8 text-slate-100 font-serif" style="text-shadow: 1px 1px 2px black">素敵な一日</p>
</div>
</div>
</div>
<div class="flex flex-col">
<section>
<div class="bg-white relative overflow-hidden">
<div class="border-4 border-orange-200 m-8 py-8 md:py-32">
<div class="mx-auto py-8 text-center">
<h2 class="text-md text-3xl felipa-regular">Message</h2>
</div>
<div class="mx-auto py-8">
<p class="text-center md:hidden">English follows Japanese.</p>
<div class="mx-auto p-8 md:p-14 max-w-2xl font-serif md:flex">
<div class="md:w-2/5 w-full">
<p class="mb-8">拝啓</p>
<p>皆様いかがお過ごしでしょうか</p>
<p>このたび結婚式を</p>
<p>執り行うこととなりました</p>
<p class="mt-4">日頃お世話になっております皆様に</p>
<p>私どもの門出を</p>
<p>お見守りいただきたく</p>
<p>ささやかながら小宴を</p>
<p>催したく存じます</p>
<p class="mt-4">ご多用中誠に恐縮ではございますが</p>
<p>ぜひご出席いただきたく</p>
<p>ご案内申し上げます</p>
<p class="my-8">敬具</p>
<p>ポアインジュン・椿原怜奈</p>
</div>
<div class="md:w-2/5 md:mt-0 mt-12 w-full md:ml-16 lg:ml-16 italic">
<p>It is with immense pleasure and gratitude that</p>
<p>Phua Yin Jun</p>
<p>and</p>
<p>Tsubakihara Reina</p>
<p>would like to invite you</p>
<p>to share a special moment of our lives.</p>
<p class="mt-4">We greatly look forward</p>
<p>to having the honor of your presence</p>
<p>as we write our own love story</p>
<p class="mt-4">We hope that</p>
<p>you will join us</p>
<p>and make our joy complete</p>
Open service 2600:9000:20ae:8800:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 62be04c57195b92a15c9e33c0bb32906.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: Q-wbyeSKmyLX4hZbAbv5WAEx1wz8Cglu14le0Rj_dPdOZSJt5446Qg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:c000:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b87d7a7588235c761c8602f922d332f4.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: tJMaKHRkVZ_VeEELs46IURBBfqyFXj3p9BPMnmOvaP-R7-wbLTelwg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:3000:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 cd703a45a77324fb8797a25a15ba227e.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: NJGbPQEN72PbcG2pwheZSguXlMuGyAwlJ1M5XA75sT233te_Z0eWdg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:b200:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 7b7e33ce27dedf9c28b39ecc0309b556.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: SY2NXmHcUsiOAkPkUTeuh6Upc8wsMn76ZgxT7RDFi9_dztz1bLi4xg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:8800:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 50a920ee7e446bd07188dda00cda68a2.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: U1lK0vNe2b2D-67-k83VOPDSL9k2ejKBfRdAh0zPXBmahDqVUblC5g==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.71:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 4b3ef7616dbf62f98d54524f0218face.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: BCDM6Md8Y4tpNAHh7yfjkwwGO7JwAaJ4Tz2y0AxMAfTkd5w9XGSo6g==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.71:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 ef17b5e16dfd912970beabcf9b8552b0.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: POu_XjL0jmuq2yFrV3oO6jDpk4DZmaRYYR14cbH_RYtD52SKAQ-gkg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:d400:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 37efbeb485d6113a0b2df63b2f651402.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: O67I3HQoQFv7G3_NWLVavqV5R9-s3nKHoDQzCrEyrme0e67l8Ybrgg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:c000:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 48bff6f682dda533442f6a9ed653d630.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: ldxKQzbn5ThY5qR0HwNj5Cd8ImfWf7HkArBBhSpJt8D0apieEM4BPA==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.63:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 7d775f7e444ed74169f0db8decde7d20.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: Sj5BkhX3mFYWLngbGnce-C7SV-iid1endtPEWM8nrEAaEDNBk2z_zw==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:6600:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 1457b39f2ccd71582289928342a87178.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 3GUcDglvnmBOfdee5_MwPaW5bpsUXtJWrq3pKRITz5dFuT0YEtt9Eg==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:3000:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 00fe48bc72383ac135425bf0b3409486.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: jpdDPKcUgmd1ezhSQtlfbZqGY1z80EvMntqIvXVdI-JHFgpuvuTlKw==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:6600:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b87d7a7588235c761c8602f922d332f4.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 1abNhX9pPjS435LvrTNiZRKH5lp39KizXSmUgn9_m-L2fGB-O6cfbw==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:b000:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 4b3ef7616dbf62f98d54524f0218face.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: Yycr9lxLmcYEhh3H4fpxinPA9hqVtRk8Xjw3I2STWgLZffXLdMIZIQ==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:b000:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 39665d11bf385fb9aabc991f857b37dc.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: R83bxgtJzMg3NzsADLXCG7o8BQ12ZcHeGF3TRVZalWKqFD5FFl5mww==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:d400:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 39665d11bf385fb9aabc991f857b37dc.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: CADTfl5NGvPbkZquXNBEGaf0heXmb527j6Z5CG7O8JZsq7Rjua-Y9g==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.99:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 e876a7ec501bf47e275a943cac96c3fe.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: IeX-TEda85cGxOAg7U9PWxpbhhFGfwv06Xb9VOqH_PLJnrm06_6ohw==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.63:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 37efbeb485d6113a0b2df63b2f651402.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: Uj2jtYxY0VTHMtwK1oesEkdFFm1nzQ8251XcR4I9v7Q51zfHzcfPnA==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.52:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2be8016001d2c9c5362b82e28629d2d6.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: yPSx5tg0lNc6y8JvDjCRghKctl8dnnZR0Yn5Lx0TlM92p6rcKd54JA==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:b200:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 ef17b5e16dfd912970beabcf9b8552b0.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: e_xnXRnx7MQcLpjk_gbA7zWFDwmdwCdrvZrxgcCVFShYc_LIrlAdtQ==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.52:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 f9e9a2e2a630392daf40b42b49debe88.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: DaNe2BcgT1DFlTrWhgUEkYJlCjYPoPUvy-bq5i6FVxzniK1tbiu7-g==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 54.230.228.99:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 8a0110b64ead65f0aff7193e350b2c52.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: dkPJbsIs1iOIRpqsGbdQPQmh-JfnYdS-DSGKtfxFVfu2CmejVM2iQg==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:9800:1c:e11a:8300:93a1:443 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b10eef4dff0375003ae9795596a9615c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 4AxpKQUgQjC5N4qyuMtmxCOZwvrDLm4X5R9QoN7RDdgrF-vVK-v9BQ==
Age: 20429
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:20ae:9800:1c:e11a:8300:93a1:80 · www.yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 65bfa9839a30709dc259dc9134cf67b2.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: MUC50-P5
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: KiClgajZXGaWC8wVt35ua_-oOhyXYuPdI5q_Nce0fsgSW0_aPfbnvw==
Age: 20428
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:e400:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 8027798dc40af04392a940303e0fc516.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: l1eg7RKAZGzHJrzga5sk3UNJPX77eHlpy3VNLF0rXSo-hnx4qtqhpA==
Age: 20399
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:6600:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b24ae08ef06aba9fb6d6ddc32eb80c64.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: yan7mVkB14-vbF7-5V7JKTWVSvmrWd5T0c1fBf5zFKKreSDtFhd3Hg==
Age: 20399
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:600:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b6b8b152d22969c98f9f56610821c954.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 5e6TMq9jtXx0Il0OEsSCiujvBe5AOLvz3ov1H_dNm-RInxsMI1roIg==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:1600:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b24ae08ef06aba9fb6d6ddc32eb80c64.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 4SVa3IsgUWKLWrR6klhU2QH1DkWRYgRnWX-unyS1rwjneJjLksgY-A==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:e400:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2d859daa66fde82c2a8685f4b0ee0dbe.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: Ps3Z8RdobQ0i-0DaqISyYJpBFtcYuGxtgC5Lr02Nk3WyOIVSgOBokw==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:8800:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2e4eab1a81a3a1decbe496056c9489da.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 3PfaOhrY7YJ9VkxxOQx-szODaCei0l2Coi5nHFdh2zYn2p8uit7Eyw==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:f600:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 df11b13b779c62601ca4cd4d2bb0ce18.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: pCV7OEFuO-8l-e26EqPk-yADmhyU-OBXi4a5OSNNoYj9RpoIKCd1Xw==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:f600:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 79006a1b1266ae3c597864512702ebf8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: mvL6mmFnWnfF3FVG_Usc9tpkJ76vJ2BJWx1onIWlAunGWJa9ZAnykA==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.62:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 1bc764620f6db84c92e8682716d68eda.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: SK1r2OwkkhIcFFoDaN9ShpMf95DRkb6vLUx3VTG5Js-X1zaMDwYi_Q==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.99:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 385cb8def78c1bb03b9aa3bd53bf1f06.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: lf2lu8QJq3w9Q9BqtZ_xhCw2Qex7AE6ddFkFE2L6cjq_xs2FBG95Uw==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:5400:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 1ff9ce989cf7cac5366389a5419a6d38.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: LD7eiO4kYfx9jXfQjC_7tZcxhvDkY5pYIYyux5caMVvN9PZ0MukcvA==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:6600:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 06659e009eac6940f260d2b396e0460c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 2qBt3q2_gTE_ZwTdRaw43klrcACJkoE2R3VM3Y2R-wBEwVtADxBlXg==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:5400:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 55c4ff21b8d72983466f815563acf5be.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 1Pl_4h9JqbBlw0Ash5rKUSOtR4h8J9aEYfI8WLZpw-tY29ndy_7a4Q==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.62:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 53fd5912708d75e6ec2b16a58625cb1e.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: SPvm6Sc2G0tk6Sm1QxlsH2APB90si7s4Lv4QowOKf4ZrNeUAJssM_A==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.99:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 1ff9ce989cf7cac5366389a5419a6d38.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: am1MkY2piOklgSdoYzXTZpyWMaBeWy_wbwSUcX0YMAZvYXAap7NqLw==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.83:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2cce65cc9ed94a08f98eec766e6667a8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: VwwCYLYsOO4H_Ncxj4rA1_Yse-FUIGkUy9xTRmrGO1uqLQ5JXqfeBg==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:600:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 203ee6b98de7af3adc87c8746659929c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: xqNaRXwrGERZ6LVjfXdY387UurgPS1kTi2upwYoTMogVznhYEsbn1w==
Age: 20399
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.11:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 7f01ac3c2b3b2aec2108ed414afd3146.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: GA2V0e9UI5TY21RIqr1PkE8atzuyNTFKotoChB6NKUc16Hrftg66Ww==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:1600:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 53fd5912708d75e6ec2b16a58625cb1e.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: Lfv6R-N5ds4w6PBs9Kr2giHWnYzfDb3crjXtnF8HSSaDGijG1jQqRA==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:f400:1c:e11a:8300:93a1:443 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 3fc67c60d4a1440649d83c01918a9054.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: DS4ZCI33Y6RRxEPzVnce7E6EfSJlXRzUkgsO9umaxw8Uo4C0lZtq3g==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.11:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 161bb6093ee10b11ad6a8a23b3138bee.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 4pnluSxdv8dBpPFilW-2KjgC-xunnCSFW3Bpk92x8QlY2oXK2-cs-A==
Age: 20398
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:8800:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 fde6fada26b7302661010feaa587bdb8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 9PmBLXbGHFW8bK3ljQVJ7peruW6TavsPI3cjKrWgzKmgpSUo3TSDxQ==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 52.222.136.83:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 32202134de03dbdd880d0da736c39c1c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: ca4mS7DsSDlMuq1UsPMQkAE5HlB9lDOQncJi1fziOxceIyQOYNJ0ig==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 2600:9000:2204:f400:1c:e11a:8300:93a1:80 · yinjunphua.com
2026-01-09 13:06
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 16885
Connection: close
Date: Fri, 09 Jan 2026 07:26:21 GMT
Last-Modified: Wed, 24 Dec 2025 04:56:56 GMT
ETag: "3ef6ffaac228d30a732239a7cf342964"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 06659e009eac6940f260d2b396e0460c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: kaK0rhx7nm7Af8W5kETGsixeFJ3RK5L1UxsAYZKqpRSDh1RLybLbBw==
Age: 20397
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, com
Open service 131.112.16.149:443 · um.yinjunphua.com
2026-01-05 02:20
HTTP/1.1 200 OK server: nginx date: Mon, 05 Jan 2026 02:20:21 GMT content-type: text/html; charset=utf-8 content-length: 11453 vary: Accept-Encoding x-dns-prefetch-control: on content-security-policy: default-src 'self'; img-src 'self' https: data:; script-src 'self' 'unsafe-eval' 'unsafe-inline'; style-src 'self' 'unsafe-inline'; connect-src 'self' https:; frame-ancestors 'self' ; vary: rsc, next-router-state-tree, next-router-prefetch, next-router-segment-prefetch, Accept-Encoding x-nextjs-cache: HIT x-nextjs-prerender: 1 x-nextjs-prerender: 1 x-nextjs-stale-time: 300 x-powered-by: Next.js cache-control: s-maxage=31536000 etag: "l9gectcv3g8u5" strict-transport-security: max-age=31536000; includeSubDomains; preload connection: close Page title: Umami <!DOCTYPE html><!--7NfGmoFbyF3nXP_KgaNk4--><html lang="en"><head><meta charSet="utf-8"/><link rel="preconnect" href="/" crossorigin=""/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="stylesheet" href="/_next/static/chunks/de0700ad3be2e209.css" data-precedence="next"/><link rel="stylesheet" href="/_next/static/chunks/a889f287f377e3e9.css" data-precedence="next"/><link rel="preload" as="script" fetchPriority="low" href="/_next/static/chunks/259ea9c815e81d24.js"/><script src="/_next/static/chunks/41ba4f39d5296fd8.js" async=""></script><script src="/_next/static/chunks/533f5ab6834b44fc.js" async=""></script><script src="/_next/static/chunks/df76c9a54ee29a76.js" async=""></script><script src="/_next/static/chunks/d57d8b0fe948f533.js" async=""></script><script src="/_next/static/chunks/turbopack-98dfdce18196de0e.js" async=""></script><script src="/_next/static/chunks/f868387fe51474b6.js" async=""></script><script src="/_next/static/chunks/b862f80a14993d06.js" async=""></script><script src="/_next/static/chunks/4b86fdb2092e3abd.js" async=""></script><script src="/_next/static/chunks/8001d5adf8a1233a.js" async=""></script><script src="/_next/static/chunks/0f14fee1bcad78c1.js" async=""></script><script src="/_next/static/chunks/11583576b68e41f6.js" async=""></script><script src="/_next/static/chunks/f54e892d6b78c6ff.js" async=""></script><script src="/_next/static/chunks/7c3248f43b5628bb.js" async=""></script><script src="/_next/static/chunks/ff2138697f233e69.js" async=""></script><link rel="icon" href="/favicon.ico"/><link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png"/><link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png"/><link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png"/><link rel="manifest" href="/site.webmanifest"/><link rel="mask-icon" href="/safari-pinned-tab.svg" color="#5bbad5"/><meta name="msapplication-TileColor" content="#da532c"/><meta name="theme-color" content="#fafafa" media="(prefers-color-scheme: light)"/><meta name="theme-color" content="#2f2f2f" media="(prefers-color-scheme: dark)"/><meta name="robots" content="noindex,nofollow"/><title>Umami</title><script src="/_next/static/chunks/a6dad97d9634a72d.js" noModule=""></script></head><body><div hidden=""><!--$--><!--/$--></div><!--$--><!--$--><!--/$--><div class="Toaster_toaster__OGJjM Toaster_position-bottom-right__MGVjY"></div><!--/$--><script src="/_next/static/chunks/259ea9c815e81d24.js" id="_R_" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0])</script><script>self.__next_f.push([1,"1:\"$Sreact.fragment\"\n2:\"$Sreact.suspense\"\n3:I[227088,[\"/_next/static/chunks/f868387fe51474b6.js\",\"/_next/static/chunks/b862f80a14993d06.js\",\"/_next/static/chunks/4b86fdb2092e3abd.js\",\"/_next/static/chunks/8001d5adf8a1233a.js\",\"/_next/static/chunks/0f14fee1bcad78c1.js\",\"/_next/static/chunks/11583576b68e41f6.js\",\"/_next/static/chunks/f54e892d6b78c6ff.js\"],\"Providers\"]\n4:I[80427,[\"/_next/static/chunks/f868387fe51474b6.js\",\"/_next/static/chunks/b862f80a14993d06.js\",\"/_next/static/chunks/4b86fdb2092e3abd.js\",\"/_next/static/chunks/8001d5adf8a1233a.js\",\"/_next/static/chunks/0f14fee1bcad78c1.js\",\"/_next/static/chunks/11583576b68e41f6.js\",\"/_next/static/chunks/f54e892d6b78c6ff.js\"],\"default\"]\n5:I[324407,[\"/_next/static/chunks/f868387fe51474b6.js\",\"/_next/static/chunks/b862f80a14993d06.js\",\"/_next/static/chunks/4b86fdb2092e3abd.js\",\"/_next/static/chunks/8001d5adf8a1233a.js\",\"/_next/static/chunks/0f14fee1bcad78c1.js\",\"/_next/static/chunks/11583576b68e41f6.js\",\"/_next/static/chunks/f54e892d6b78c6ff.js\"],\"default\"]\n6:I[329306,[\"/_next/static/chunks/f868387fe51474b6.js\",\"/_next/static/chunks/b862f80a14993d06.js\",\"/_next/static/chunks/4b86fdb2092e3abd.js\",\"/_next/static/chunks/8001d5adf8a1233a.js\",\"/_next/static/chunks/0f14fee1bcad78c1.js\",\"/_next/static/chunks/11583576b68e41f6.js\",\"/_next/static/chunks/f54e892d6b78c6ff.js\",\"/_next/static/chunks/7c3248f43b5628bb.js\"],\"default\"]\n7:I[299705,[\"/_next/stat
Open service 131.112.16.149:80 · um.yinjunphua.com
2026-01-05 02:20
HTTP/1.1 301 Moved Permanently content-length: 0 location: https://um.yinjunphua.com/ connection: close
Open service 2600:9000:2204:d400:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 d11dcb69d5fad0d00cf6f2c45df4bb94.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: prRMJHr1OLojCD2GRmkwjM49WBCg7oUwhZ8jbczFSUyG7xaGFLGltQ==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.11:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2696c49ebf3abec704c6af790acf6778.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 5_jOYwsY8mJNGX-0QtjtEF4v5y-KVne53W0Em2daRikXjNAGAYX8cg==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.99:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 d3567cb7a3a02c495b66e54c187b0ea2.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: XuYaXjEu81vJ32DEBSJLj7d3svZSloiCX1XoxDyvLYhmRIMQwi927A==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.83:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 4f762327597c2647c5dac5e573d910fa.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: OZA-chwXRyp8m--UcV4ZOg4Ym-c6YOdio6sXB-mY8F9nhVSAOORkcQ==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.62:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 1e882280b9c5d046c63d8cd0c1faf9c0.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: BuqYVrpF-dI0fp2wtv1JUItAaZ6wQB6DbZJgFOxO4RGgEMXZebkj5A==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:a200:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 df11b13b779c62601ca4cd4d2bb0ce18.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: hqA80oVLvxPbEIKjvoVA730nMKSd-2G-87Y1WPvIiJh-zmA_Ro0wwA==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:6a00:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 1ad022b197e464938f7729463478f0c8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: tVUC0-p0CGxA7yaeylecPp4jFS498wiGKQ05MmT_mfmJrkK19BlROw==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:7600:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 fde6fada26b7302661010feaa587bdb8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: ZAYgWVcXQNt5VqEkXdVDGyVEaTCCSoLlMoscqWxWvgkNS0ywGbxDBA==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.11:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 d3567cb7a3a02c495b66e54c187b0ea2.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: aIWC3mbUYdXW1QjmbL8LNtIiRIvBXwf8e5f8Sk1qN_QT0hHmf1ooUg==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.99:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 d11dcb69d5fad0d00cf6f2c45df4bb94.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: dOPpJDiR5J-7-O_gMZRlfD4Odp2V7WVk-XyslscV7Mnaxw5qRstnEA==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:a200:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 7f01ac3c2b3b2aec2108ed414afd3146.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: lDp0cPR1-kz4Nr8yL70T5Ww4abZmvK27T-nmNQJXhfGcczCGAmUG9A==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.83:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 8027798dc40af04392a940303e0fc516.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 2ScRZJqOCt7nmJ9KE7mOqugmcyQX5rwAYvIacS9IMFpSTHs_E-B98g==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:d400:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 79006a1b1266ae3c597864512702ebf8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 340qOpCfuSVueeqfZJuyZMSmFDvkw0r6ZUrbzQ0rkPTIjlwFpQOqeQ==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:3a00:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 89d30ce8a4c37b9d11d7f552521193ae.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: pOR_hRU16RxorXegU9Y2T8ZWw7allmhZUEpSauGpBnYjEw-3maijug==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:5e00:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 d3567cb7a3a02c495b66e54c187b0ea2.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: oswJe7owYG0n7MQv2m2aQxT5gyvn8IoVxOPh5_MWYRMTd1UhihCX_g==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:c400:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 32202134de03dbdd880d0da736c39c1c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: XVIYpkaG7P296-BQl6cxkYybDvLKl5ra5ko1WE1ZYl2JckPRqSWYqg==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:d200:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 161bb6093ee10b11ad6a8a23b3138bee.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: -t-PJpcQZAoTdp9nKx0Wjy6GfNsFnQjpCpybhroLHe6mEo9a470dRA==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:3a00:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 385cb8def78c1bb03b9aa3bd53bf1f06.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: epVyuXePLJn2cCZiQ7rijfEtf7QMmLJuX0AwbZt9YhkBPAv38G0Hvg==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:5e00:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 b24ae08ef06aba9fb6d6ddc32eb80c64.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: g89xK4hT_nEVT4-rz_9xQ7vD-v7l9c-5j2EdOV2ThJ0jCQVSt1_VAw==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 52.222.136.62:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 32202134de03dbdd880d0da736c39c1c.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: ViFPpegcSbFMn3GG6TOiqS2zXQ_h-N8kR3gQcso03EViMOWEbS4Nng==
Age: 58302
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:7600:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 161bb6093ee10b11ad6a8a23b3138bee.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: lmkDUZhxuFvkOU6Ywbl65zr-vpsSpiJt5ttUqOGzXDX0RHQjWsP2Vg==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:d200:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2cce65cc9ed94a08f98eec766e6667a8.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: jiDW9kgsXOF-dEtk6IcWmkQoCnZwY__72rJGI4tDlDeik3d8ZlZaqQ==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:6a00:1c:e11a:8300:93a1:443 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 f353b9615396320dcfec689a26cf519e.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: 1y7rZYZmL7ZXEtU81cX4a217LCFygjzaI2B9W_AQDfe99f7sHy0OJg==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 2600:9000:2204:c400:1c:e11a:8300:93a1:80 · yinjunphua.com
2025-12-19 10:34
HTTP/1.1 200 OK
Content-Type: text/html
Content-Length: 17051
Connection: close
Date: Thu, 18 Dec 2025 18:23:15 GMT
Last-Modified: Mon, 08 Dec 2025 07:57:43 GMT
ETag: "bbdbfb71804cb08499afe6252d393e03"
Accept-Ranges: bytes
Server: web
Vary: Accept-Encoding
X-Cache: Hit from cloudfront
Via: 1.1 2d859daa66fde82c2a8685f4b0ee0dbe.cloudfront.net (CloudFront)
X-Amz-Cf-Pop: FRA50-P2
Alt-Svc: h3=":443"; ma=86400
X-Amz-Cf-Id: I6AgXtRFShsKhfpoQWciK1U8D00ZP0Ac3ZHZ14pPfHz1kJRXEdncXw==
Age: 58301
Page title: Yin Jun, Phua -- Assistant Professor at Science Tokyo
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Yin Jun, Phua -- Assistant Professor at Science Tokyo</title>
<meta name="description" content="Yin Jun Phua, currently an assistant professor at Institute of Science Tokyo. Main research focus is on bridging the gap between symbolic AI and neural networks.">
<link rel="stylesheet" href="/css/chota.css">
<link rel="stylesheet" href="/css/style.css">
<link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
<link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
<link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png">
<link rel="manifest" href="/site.webmanifest">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="alternate" hreflang="en" href="https://yinjunphua.com/">
<link rel="alternate" hreflang="ja" href="https://yinjunphua.com/ja/">
<link rel="canonical" href="https://yinjunphua.com/">
<meta http-equiv="content-language" content="en">
<script defer src="https://um.yinjunphua.com/script.js" data-website-id="553526c2-8bba-4565-af3b-38ce4c9e4002"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col">
<div class="pull-right">
EN | <a href="/ja/">日本語</a>
</div>
</div>
</div>
<nav class="nav">
<div class="nav-left">
<h1 class="is-marginless">Yin Jun, Phua <small>— Ph.D. in Informatics</small></h1>
</div>
<div class="nav-right">
<a href="/my-research.html">My Research</a>
</div>
</nav>
<hr>
<div class="row">
<div class="col-10">
<p><span class="text-grey">Currently</span> <strong>Assistant Professor</strong> <span class="text-grey">at</span> Institute of Science Tokyo <span class="text-grey">since</span> October, 2024.</p>
<p><span class="text-grey">My main research focus is on</span> bridging the gap between <strong>symbolic AI</strong> and <strong>neural networks</strong>.</p>
<p class="text-grey">During my Ph.D., I focused on developing deep learning techniques that could extract symbolic rules from time series data. The symbolic rules generated by these techniques were meant to serve as an explanation for the observed data, and could be used for prediction and manipulation of dynamic systems. My work aimed to create explainable AI systems that could provide human-accessible insights into complex systems.</p>
<p class="text-grey">Currently, I am exploring a new technique that combines the strengths of both symbolic and neural networks. While symbolic AI provides insights that are directly transferable and learnable by humans, its brittleness limits its real-world application. On the other hand, neural networks are robust against noise and have broad real-world applications, but the knowledge and insights they provide can be difficult to access. By combining both approaches, I hope to devise intelligent systems that are both robust and accessible to humans. My goal is to create explainable AI systems that can provide valuable insights into complex systems while also being practical for real-world applications.</p>
</div>
<div class="col-2">
<picture>
<source srcset="/images/profile_2023.webp" type="image/webp">
<source srcset="/images/profile_2023.png" type="image/png">
<img src="/images/profile_2023.png" alt="My Profile Picture 2023">
</picture>
</div>
</div>
<p><span class="text-grey">Research areas: </span>deep learning, symbolic rule extraction, time series analysis, explainable AI, hybrid approach, neural networks, robustness, and human-accessible insights.</p>
<p class="text-grey">Have interesting projects or want to have some discussion? You can contact me at <span class="text-dark">phua<span class="replace"> [at] </span>comp.isct.ac.jp</span></p>
<h3>Goal</h3>
<p class="text-grey">My main focus is on bridging the gap between symbolic AI and neural networks, combining the strengths of both approaches to create more <strong class="text-dark">robust</strong> and <strong class="text-dark">flexi
Open service 131.112.16.149:80 · container.yinjunphua.com
2025-12-19 09:05
HTTP/1.1 301 Moved Permanently content-length: 0 location: https://container.yinjunphua.com/ connection: close
Open service 131.112.16.149:443 · container.yinjunphua.com
2025-12-19 09:05
HTTP/1.1 401 Unauthorized content-length: 112 cache-control: no-cache content-type: text/html www-authenticate: Basic realm="Authentication" connection: close <html><body><h1>401 Unauthorized</h1> You need a valid user and password to access this content. </body></html>
131.112.16.149 4 54.230.228.63 1 2600:9000:2204:1600:1c:e11a:8300:93a1 1 2600:9000:20ae:8800:1c:e11a:8300:93a1 1 2600:9000:20ae:c000:1c:e11a:8300:93a1 1 2600:9000:20ae:3000:1c:e11a:8300:93a1 1 2600:9000:20ae:b200:1c:e11a:8300:93a1 1 54.230.228.71 1 2600:9000:20ae:d400:1c:e11a:8300:93a1 1 2a06:98c1:3121::3 1 2600:9000:20ae:6600:1c:e11a:8300:93a1 1 2600:9000:20ae:b000:1c:e11a:8300:93a1 1 54.230.228.99 1 54.230.228.52 1 2600:9000:20ae:9800:1c:e11a:8300:93a1 1 2600:9000:2204:e400:1c:e11a:8300:93a1 1 2600:9000:2204:6600:1c:e11a:8300:93a1 1 2a06:98c1:3120::3 1 2600:9000:2204:600:1c:e11a:8300:93a1 1 2600:9000:2204:8800:1c:e11a:8300:93a1 1 2600:9000:2204:f600:1c:e11a:8300:93a1 1 52.222.136.62 1 52.222.136.99 1 2600:9000:2204:5400:1c:e11a:8300:93a1 1 52.222.136.83 1 52.222.136.11 1 2600:9000:2204:f400:1c:e11a:8300:93a1 1 188.114.97.3 1 2600:9000:2204:d400:1c:e11a:8300:93a1 1 2600:9000:2204:a200:1c:e11a:8300:93a1 1 2600:9000:2204:6a00:1c:e11a:8300:93a1 1 2600:9000:2204:7600:1c:e11a:8300:93a1 1 2600:9000:2204:3a00:1c:e11a:8300:93a1 1 2600:9000:2204:5e00:1c:e11a:8300:93a1 1 2600:9000:2204:c400:1c:e11a:8300:93a1 1 2600:9000:2204:d200:1c:e11a:8300:93a1 1