FRE Special Seminar: Luciano Campi

Luciano Campi

Title

Optimal Coarse Correlated Equilibria in Mean Field Games

Abstract

We study coarse correlated equilibria (CCEs) in continuous-time mean field games. CCEs generalize Nash equilibria by allowing a moderator, or correlation device, to randomize over strategies so that no player can profitably deviate unilaterally before observing the recommendation. We develop a linear programming approach based on relaxed strategies, following the spirit of Kurtz and Stockbridge and recent extensions to mean field games by Bouveret, Dumitrescu, Leutscher, and Tankov. In this relaxed framework, under suitable regularity assumptions, we prove the existence of an optimal CCE for a fixed moderator objective. We also derive an equivalent Lagrangian formulation and propose a primal-dual algorithm for computing optimal CCEs numerically. This talk is based on joint work with F. Cannerozzi and I. Tzouanas.

Bio

Luciano Campi is a Full Professor of Probability and Mathematical Statistics in the Department of Mathematics “Federigo Enriques” at the University of Milan. Previously, he was Associate Professor in the Department of Statistics at the London School of Economics and Political Science (LSE), Professor of Mathematics at the University Paris 13, and Assistant Professor at University Paris Dauphine. Luciano holds a PhD in Mathematics from the University of Paris 6 and in Computational Mathematics from the University of Padua. His research focuses on stochastic control and stochastic differential games, mean field games, and applications to energy markets, information asymmetry, and market frictions.