BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:From Predictions to Decisions in Dynamic Strategic Environments
DTSTART:20260205T111500
DTEND:20260205T120000
DTSTAMP:20260918T123655Z
UID:891281646ee63eb48f75ec5581e41bdf0eaa98eee33c39e1138590bd
CATEGORIES:Conferences - Seminars
DESCRIPTION:Aymeric Capitaine\, PhD student at École Polytechnique/Inria 
 Paris\, France\nAbstract:\nMany real-world games take place in environment
 s that change over time due to external and sometimes adversarial shocks. 
 Adapting strategies to these changes can be difficult for players. However
 \, they often have access to predictions of future states of nature\, for 
 example from machine learning models. This presentation studies how such p
 redictions can be used in non-stationary strategic environments. First\, w
 e introduce prediction-aware learning\, a game-theoretic framework for pla
 yers who rely on forecasts in time-varying games. We present learning algo
 rithms that use predictions and provide guarantees on equilibrium converge
 nce and social welfare. Second\, we study how players should train their p
 rediction models. Simply minimizing a standard prediction error is not alw
 ays the best approach\, since accurate predictions do not necessarily lead
  to good actions. This motivates decision-focused learning\, where models 
 are trained to directly improve the quality of the resulting decisions. We
  introduce an online decision-focused learning framework\, along with algo
 rithms\, theoretical guarantees\, and experiments. We illustrate these ide
 as with an application to power markets.\n\nBiography:\nAymeric Capitaine 
 is a third-year PhD student at École Polytechnique/Inria Paris\, supervis
 ed by Michael I. Jordan\, Alain Durmus\, and Étienne Boursier. His resear
 ch focuses on the strategic foundations of multi-agent systems. He revisit
 s classical microeconomic models through the lens of modern learning metho
 ds and studies the interaction between prediction and decision-making in t
 ime-varying environments.\n 
LOCATION:ME C2 405 https://plan.epfl.ch/?room==ME%20C2%20405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR