BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Tutorial: Generalization in Reinforcement Learning: From Foundatio
 ns to New Frontiers - Csaba  Szepesvári (University of Alberta)
DTSTART:20251105T100000Z
DTEND:20251105T130000Z
UID:TALK235618@talks.cam.ac.uk
DESCRIPTION:Reinforcement learning (RL) and optimal control share a deep i
 ntellectual heritage in addressing sequential decision-making under uncert
 ainty. This tutorial develops a computer scientist&rsquo\;s&nbsp\;perspect
 ive on RL theory&mdash\;one that places generalization\, sample efficiency
 \, and computational tractability at the center of the analysis. A particu
 lar focus will be on the stylized setting of linear&nbsp\;function approxi
 mation\, which offers the best prospects for developing and understanding 
 tractable algorithms. The tutorial will illustrate how this perspective sh
 apes problem formulations\,&nbsp\;abstractions\, and algorithmic insights 
 through several representative results. It will conclude by considering ho
 w similar ideas might inform reasoning and planning in large language mode
 ls\, raising&nbsp\;more questions than answers.\n&nbsp\;\nThe tutorial fol
 lows the new MIT Press textbook "Multi-Agent Reinforcement Learning: Found
 ations and Modern Approaches"\, available at www.marl-book.com.\n&nbsp\;
LOCATION:Enigma Room\, The Alan Turing Institute
END:VEVENT
END:VCALENDAR