BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//NextToU//EN
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:055562d5-b6e7-4ddf-92af-e2e02448e07f@nexttoyou
DTSTAMP:20261106T163859Z
DTSTART:20261120T163000Z
DTEND:20261120T173000Z
SUMMARY:CIS Seminar Series 2026 - Haimin Hu\, PhD
LOCATION:Kirkbride Lecture Hall 204\, Kirkbride Lecture Hall\, University 
 of Delaware\, Newark\, DE 19716\, USA
DESCRIPTION:Game-theoretic safety and agility for human-centered robotics\
 n\nABSTRACT\nFrom autonomous vehicles navigating busy intersections to mo
 bile manipulators deployed in household environments\, robots must operat
 e safely and efficiently around people in uncertain and unstructured situ
 ations. Yet today’s robots still struggle to robustly handle human-induce
 d uncertainty without sacrificing efficiency when working closely with pe
 ople. This talk will argue that jointly reasoning the robots' physical co
 ntexts and belief states that capture human uncertainty offers a unifying
  lens for trustworthy human–robot interaction (HRI). We begin by introduc
 ing a belief-space safety filtering framework that combines robust safety
  analysis with probabilistic inference to enable assured interactions acr
 oss diverse HRI scenarios. We will then show how such filters can be scal
 ably synthesized via game-theoretic reinforcement learning and statistica
 lly verified using conformal prediction. Finally\, we present a belief-sp
 ace dynamic game in which a robot coach helps a human learner acquire and
  retain motor skills.\n\nBIOGRAPHY\nHaimin Hu is an assistant professor o
 f computer science and a core member of the Johns Hopkins Data Science an
 d AI Institute\, the Laboratory for Computational Sensing and Robotics\, 
 and the Institute for Assured Autonomy. His research focuses on the algor
 ithmic foundations of human-centered autonomy. By integrating dynamic gam
 e theory with machine learning and safety-critical control\, h\n\nhttp://
 next-to-u.app/events/055562d5-b6e7-4ddf-92af-e2e02448e07f
URL:http://next-to-u.app/events/055562d5-b6e7-4ddf-92af-e2e02448e07f
END:VEVENT
END:VCALENDAR
