BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UM//UM*Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Detroit
TZURL:http://tzurl.org/zoneinfo/America/Detroit
X-LIC-LOCATION:America/Detroit
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20070311T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20071104T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250218T140613
DTSTART;TZID=America/Detroit:20250416T160000
DTEND;TZID=America/Detroit:20250416T170000
SUMMARY:Workshop / Seminar:Optimal PhiBE — A Model-Free PDE-Based Framework for Continuous-Time Reinforcement Learning
DESCRIPTION:This talk addresses continuous-time reinforcement learning (RL) in settings where the system dynamics are governed by a stochastic differential equation but remains unknown\, with only discrete-time observations available. While the optimal Bellman equation (optimal-BE) enables model-free algorithms\, its discretization error is significant when the reward function oscillates. Conversely\, model-based PDE approaches offer better accuracy but suffer from non-identifiable inverse problems. \nTo bridge this gap\, we introduce Optimal-PhiBE\, an equation that integrates discrete-time information into a PDE\, combining the strengths of both RL and PDE formulations. Compared to the RL formulation\, Optimal-PhiBE is less sensitive to reward oscillations\, leading to smaller discretization errors. In linear-quadratic control\, Optimal-PhiBE can even achieve accurate continuous-time optimal policy with only discrete-time information. Compared to the PDE formulation\, it skips the identification of the dynamics and enables model-free algorithm derivation. Furthermore\, we extend Optimal-PhiBE to higher orders\, providing increasingly accurate approximations.
UID:129844-21864640@events.umich.edu
URL:https://events.umich.edu/event/129844
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Mathematics
LOCATION:East Hall - 1360
CONTACT:
END:VEVENT
END:VCALENDAR