BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Learning equilibria in games with zeroth-order information
DTSTART:20230516T161500
DTEND:20230516T171500
DTSTAMP:20260508T175201Z
UID:de6f138d6d63f44dcde5ba2d007b0b40b377fc7d24a9d6ee2b32c3ef
CATEGORIES:Conferences - Seminars
DESCRIPTION:Maryam Kamgarpour (EPFL)\nI address the question of learning e
 quilibria in convex and non-convex games under zeroth-order information. I
 n the convex setting\, I will present our learning algorithm for monotone 
 games that leverages estimation of game pseudo-gradient. I discuss extensi
 ons of the approach to a broader class of games as well as its convergence
  rates. In the non-convex setting\, I will present our no-regret algorithm
  that leverages probabilistic estimation of a players' cost function. I wi
 ll compare the regret rate of the algorithm with the state-of-the-art and 
 discuss our extensions to multi agent reinforcement learning.\n\n 
LOCATION:GA 3 21 https://plan.epfl.ch/?room==GA%203%2021
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR