A General Framework for Learning Mean-Field Games
From MaRDI portal
Abstract: This paper presents a general mean-field game (GMFG) framework for simultaneous learning and decision-making in stochastic games with a large population. It first establishes the existence of a unique Nash Equilibrium to this GMFG, and demonstrates that naively combining reinforcement learning with the fixed-point approach in classical MFGs yields unstable algorithms. It then proposes value-based and policy-based reinforcement learning algorithms (GMF-V and GMF-P, respectively) with smoothed policies, with analysis of their convergence properties and computational complexities. Experiments on an equilibrium product pricing problem demonstrate that GMF-V-Q and GMF-P-TRPO, two specific instantiations of GMF-V and GMF-P, respectively, with Q-learning and TRPO, are both efficient and robust in the GMFG setting. Moreover, their performance is superior in convergence speed, accuracy, and stability when compared with existing algorithms for multi-agent reinforcement learning in the -player setting.
Recommendations
- Unified reinforcement Q-learning for mean field game and control problems
- Actor-critic reinforcement learning algorithms for mean field games in continuous time, state and action spaces
- Model-free mean-field reinforcement learning: mean-field MDP and mean-field Q-learning
- Learning in mean field games: The fictitious play
- Reinforcement learning for non-stationary discrete-time linear-quadratic mean-field games in multiple populations
Cited in
(11)- Generalized conditional gradient and learning in potential mean field games
- Graphon mean-field control for cooperative multi-agent reinforcement learning
- Model-free mean-field reinforcement learning: mean-field MDP and mean-field Q-learning
- Learning Optimal Policies in Potential Mean Field Games: Smoothed Policy Iteration Algorithms
- MF-OMO: An Optimization Formulation of Mean-Field Games
- Actor-critic reinforcement learning algorithms for mean field games in continuous time, state and action spaces
- Data-driven stability of stochastic mean-field type games via noncooperative neural network adversarial training
- On the effect of time preferences on the price of anarchy
- MF-OML: online mean-field reinforcement learning with occupation measures for large population games
- On time-inconsistency in mean-field games
- Recent progress in finite-population games: mean field game-based approaches
This page was built for publication: A General Framework for Learning Mean-Field Games
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6199266)
