Maryam Kamgarpour: Learning equilibria in multiagent systems with bandit feedback

A significant challenge in managing large-scale engineering systems, such as energy and transportation networks, lies in enabling autonomous decision-making of interacting agents. Game theory offers a framework for modeling and analyzing this class of problems. In many practical applications, each player only has partial information about the cost functions and actions of others. Therefore, a decentralized learning approach is essential to devise optimal strategies for each player.

My talk will focus on recent advances in decentralized learning in static and in Markov games under bandit feedback. It highlights challenges compared to single agent learning and presents algorithms with provable convergence. The first part will focus on learning in continuous action static games. The second part will address Markov games, presenting our learning approaches for zero-sum Markov games and coarse-correlated equilibria in general-sum Markov games. I will highlight a few applications of the proposed methods and will conclude with few open research directions.