BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//https://caida.ubc.ca//NONSGML iCalcreator 2.41.92//
CALSCALE:GREGORIAN
METHOD:PUBLISH
UID:63383937-3436-4232-b532-646661646364
X-WR-RELCALID:efc09d74-9c93-479e-a94f-485231ddccde
X-WR-TIMEZONE:America/Vancouver
X-WR-CALNAME:Reinforcement Learning With Constraints: From Theory to Reason
 ing in LLM - Lin Yang\, Assistant Professor\, UCLA
BEGIN:VTIMEZONE
TZID:America/Vancouver
BEGIN:STANDARD
TZNAME:PST
DTSTART:20241103T020000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RDATE:20251102T020000
END:STANDARD
BEGIN:STANDARD
TZNAME:MST
DTSTART:20261101T020000
TZOFFSETFROM:-0700
TZOFFSETTO:-0700
END:STANDARD
BEGIN:DAYLIGHT
TZNAME:PDT
DTSTART:20250309T020000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RDATE:20260308T020000
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:3953aff3-12eb-4d2b-97ca-6441b38a9758
DTSTAMP:20260824T005443Z
CLASS:PUBLIC
CREATED:20250710T211821Z
DESCRIPTION:Abstract: In this talk\, I will explore reinforcement learning 
 with constraints\, focusing on both theoretical foundations and practical 
 applications. I will first present recent advances in the sample complexit
 y of constrained Markov decision processes (CMDPs)\, covering both offline
  and online settings. Our results establish near-optimal upper and lower b
 ounds under relaxed and strict feasibility regimes\, revealing that constr
 aint satisfaction—while generally harder—can match the sample efficiency o
 f unconstrained MDPs under certain conditions. These insights are grounded
  in primal-dual…
DTSTART;TZID=America/Vancouver:20250715T144500
DTEND;TZID=America/Vancouver:20250715T154500
LAST-MODIFIED:20250715T152032Z
LOCATION:UBC Vancouver Campus\, Fried Kaiser (KAIS) building\, Room 2020/20
 30\, 2332 Main Mall
SUMMARY:Reinforcement Learning With Constraints: From Theory to Reasoning i
 n LLM - Lin Yang\, Assistant Professor\, UCLA
TRANSP:OPAQUE
URL:https://caida.ubc.ca/index.php/event/reinforcement-learning-constraints
 -theory-reasoning-llm-lin-yang-assistant-professor-ucla
END:VEVENT
END:VCALENDAR