BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Europe/London BEGIN:DAYLIGHT DTSTART:20260329T020000 TZOFFSETFROM:+0000 TZOFFSETTO:+0100 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3 TZNAME:BST END:DAYLIGHT BEGIN:STANDARD DTSTART:20261025T010000 TZOFFSETFROM:+0100 TZOFFSETTO:+0000 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:GMT END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260924T230410Z UID:0B016D0F-92F6-4B66-A00E-E8530EC6C182 DTSTART;TZID=Europe/London:20261021T110000 DTEND;TZID=Europe/London:20261021T123000 DESCRIPTION:Prof. Jun Tani (Okinawa Institute of Science and Technology\, J apan)\n\nBio: Jun Tani received the D.Eng. degree from Sophia University\, Tokyo in 1995. He started his research career with Sony Computer Science Lab. in 1993. He became a PI in RIKEN Brain Science Institute in 2001. He became a tenured Professor at KAIST\, South Korea in 2012. He is currently a full Professor at OIST. He is also a visiting professor at The Technica l University of Munich. His current research interests include cognitive n euroscience\, developmental psychology\, phenomenology\, complex adaptive systems\, and robotics. He is an author of “Exploring Robotic Minds: Act ions\, Symbols\, and Consciousness as Self-Organizing Dynamic Phenomena." published from Oxford Univ. Press in 2016.\n\nAbstract: How can infants de velop skills for action and language from relatively limited experience th rough interaction with the world? This seminar presents a developmental ro botics perspective on this question by investigating how compositionality —the capacity to systematically recombine known concepts in novel situat ions—can emerge through embodied interaction. First\, we introduce a bra in-inspired neural model based on predictive coding\, active inference\, a nd the free-energy principle\, in which language\, vision\, and action are learned jointly [1]. Robotic experiments demonstrate that compositional l inguistic representations emerge through sensorimotor experience\, enablin g generalization to previously unseen language-action combinations. Second \, we discuss how self-exploration and intrinsic motivation can further su pport the development of systematic generalization under sparse learning c onditions [2]. By integrating active inference with reinforcement learning \, robots learn not only to achieve goals but also to seek information and behavioral novelty. The results suggest that richer compositional experie nces promote the emergence of structured internal representations that sup port generalization beyond training examples. Together\, these studies sug gest that compositional intelligence can emerge from embodied predictive l earning and active exploration\, offering a possible developmental account of how systematic generalization can arise from relatively limited experi ence.\n[1] Vijayaraghavan\, P.\, Queißer\, J. F.\, Verduzco-Flores\, S.\, & Tani\, J. “Development of compositionality through interactive learni ng of language and action of robots.” Science Robotics\, 10\, eadp0751\, 2025.\n[2] Tinker\, T. J.\, Doya\, K.\, & Tani\, J. “Curiosity-Driven D evelopment of Action and Language in Robots Through Self-Exploration.” S cience Advances\, 12\, eaee7533.\n\nAgenda: \nProf. Jun Tani from the Okin awa Institute of Science and Technology\, Japan\, will present a Prestigio us Lecture on "From Embodied Language Grounding to Systematic Generalizati on: Developmental Origins of Compositionality in Robots" at 12:00 (UTC+1) on Oct 21st 2026.\n\nThe talk will last 60 minutes distributed between the Prestigious Lecture\, followed by questions & answers.\n\nVirtual: https: //events.vtools.ieee.org/m/579768 LOCATION:Virtual: https://events.vtools.ieee.org/m/579768 ORGANIZER:figueredo@ieee.org SEQUENCE:12 SUMMARY:[IEEE UKI-RAS]-[Prestigious Talk] Prof. Jun Tani --- From Embodied Language Grounding to Systematic Generalization: Developmental Origins of Compositionality in Robots URL;VALUE=URI:https://events.vtools.ieee.org/m/579768 X-ALT-DESC:Description: <br /><p><strong>Prof. Jun Tani&nbsp\; (</strong>Ok inawa Institute of Science and Technology\, Japan<strong>)</strong></p>\n< p><strong>Bio: &nbsp\; </strong>Jun Tani received the D.Eng. degree from S ophia University\, Tokyo in 1995. He started his research career with Sony Computer Science Lab. in 1993. He became a PI in RIKEN Brain Science Inst itute in 2001. He became a tenured Professor at KAIST\, South Korea in 201 2. He is currently a full Professor at OIST. He is also a visiting profess or at The Technical University of Munich. His current research interests i nclude cognitive neuroscience\, developmental psychology\, phenomenology\, complex adaptive systems\, and robotics. He is an author of &ldquo\;Explo ring Robotic Minds: Actions\, Symbols\, and Consciousness as Self-Organizi ng Dynamic Phenomena." published from Oxford Univ. Press in 2016.</p>\n<p> <strong>Abstract</strong>: <span style="background-color: white\;">How can infants develop skills for action and language from relatively limited ex perience through interaction with the world? This seminar presents a devel opmental robotics perspective on this question by investigating how compos itionality&mdash\;the capacity to systematically recombine known concepts in novel situations&mdash\;can emerge through embodied interaction. First\ , we introduce a brain-inspired neural model based on predictive coding\, active inference\, and the free-energy principle\, in which language\, vis ion\, and action are learned jointly [1]. Robotic experiments demonstrate that compositional linguistic representations emerge through sensorimotor experience\, enabling generalization to previously unseen language-action combinations. Second\, we discuss how self-exploration and intrinsic motiv ation can further support the development of systematic generalization und er sparse learning conditions [2]. By integrating active inference with re inforcement learning\, robots learn not only to achieve goals but also to seek information and behavioral novelty. The results suggest that richer c ompositional experiences promote the emergence of structured internal repr esentations that support generalization beyond training examples. Together \, these studies suggest that compositional intelligence can emerge from e mbodied predictive learning and active exploration\, offering a possible d evelopmental account of how systematic generalization can arise from relat ively limited experience.</span></p>\n<div style="text-align: left\; text- indent: 0px\; background-color: white\; margin: 1em 0px\; font-family: Apt os\,Aptos_EmbeddedFont\,Aptos_MSFontService\,Calibri\,Helvetica\,sans-seri f\; font-size: 11pt\; color: black\;">[1] Vijayaraghavan\, P.\, Quei&szlig \;er\, J. F.\, Verduzco-Flores\, S.\, &amp\; Tani\, J. &ldquo\;Development of compositionality through interactive learning of language and action o f robots.&rdquo\; <em>Science Robotics</em>\, 10\, eadp0751\, 2025.</div>\ n<div style="text-align: left\; text-indent: 0px\; background-color: white \; margin: 1em 0px\; font-family: Aptos\,Aptos_EmbeddedFont\,Aptos_MSFontS ervice\,Calibri\,Helvetica\,sans-serif\; font-size: 11pt\; color: black\;" >[2] Tinker\, T. J.\, Doya\, K.\, &amp\; Tani\, J. &ldquo\;Curiosity-Drive n Development of Action and Language in Robots Through Self-Exploration.&r dquo\; <em>Science Advances</em>\, 12\, eaee7533.</div>\n<div style="text- align: left\; text-indent: 0px\; background-color: white\; margin: 1em 0px \; font-family: Aptos\,Aptos_EmbeddedFont\,Aptos_MSFontService\,Calibri\,H elvetica\,sans-serif\; font-size: 11pt\; color: black\;">&nbsp\;</div>\n<p >&nbsp\;</p><br /><br />Agenda: <br /><p><strong>&nbsp\;Prof. Jun Tani</st rong> from the Okinawa Institute of Science and Technology\, Japan\,&nbsp\ ; will present a Prestigious Lecture on "<span style="color: rgb(0\,0\,0)\ ;"><strong>From Embodied Language Grounding to Systematic Generalization: Developmental Origins of Compositionality in Robots</strong></span>" at 12 :00 (UTC+1) on Oct 21st 2026.&nbsp\;</p>\n<p>The talk will last 60 minutes distributed between the Prestigious Lecture\, followed by questions &amp\ ; answers.&nbsp\;</p> END:VEVENT END:VCALENDAR