Jimin Lin

Title: Reinforcement Learning for Mean Field Control Games

Abstract: We present a new combined mean field control game (MFCG) problem which can be interpreted as a competitive game between collaborating groups and its solution as a Nash equilibrium between the groups. Within each group the players coordinate their strategies. A three-timescale reinforcement learning algorithm is designed to approximate the solution of such MFCG problems. We apply the MFCG framework to the classical trader's problem and test the algorithm on a benchmark linear-quadratic specification for which we have an analytic solution.


Joint work with A. Angiuli, N. Detering, JP Fouque, and M. Laurière