Generalized conditional gradient and learning in potential mean field games

Abstract

We apply the generalized conditional gradient algorithm to potential mean field games and we show its well-posedeness. It turns out that this method can be interpreted as a learning method called fictitious play. More precisely, each step of the generalized conditional gradient method amounts to compute the best-response of the representative agent, for a predicted value of the coupling terms of the game. We show that for the learning sequence ${\delta }_k=2/(k+2)$, the potential cost converges in $O(1/k)$, the exploitability and the variables of the problem (distribution, congestion, price, value function and control terms) converge in $O(1/\sqrt{k})$, for specific norms.

References in corpus (10)

• A probabilistic weak formulation of mean field games and applications
• Electrical Vehicles in the Smart Grid: A Mean Field Game Analysis
• Finite mean field games: fictitious play and convergence to a first order continuous mean field game
• Learning in anonymous nonatomic games with applications to first-order mean field games
• Scaling up Mean Field Games with Online Mirror Descent
• Concave Utility Reinforcement Learning: the Mean-Field Game Viewpoint
• Generalized conditional gradient: analysis of convergence and applications
• Global-in-time regularity via duality for congestion-penalized Mean Field Games
• Exploration noise for learning linear-quadratic mean field games
• Convergence Analysis of the Frank-Wolfe Algorithm and Its Generalization in Banach Spaces