Q-learning in regularized mean-field games
Date
2022-05-23Source Title
Dynamic Games and Applications
Print ISSN
2153-0785
Electronic ISSN
2153-0793
Publisher
Birkhaeuser Science
Pages
1 - 29
Language
English
Type
ArticleItem Usage Stats
7
views
views
2
downloads
downloads
Abstract
In this paper, we introduce a regularized mean-field game and study learning of this game under an infinite-horizon discounted reward function. Regularization is introduced by adding a strongly concave regularization function to the one-stage reward function in the classical mean-field game model. We establish a value iteration based learning algorithm to this regularized mean-field game using fitted Q-learning. The regularization term in general makes reinforcement learning algorithm more robust to the system components. Moreover, it enables us to establish error analysis of the learning algorithm without imposing restrictive convexity assumptions on the system components, which are needed in the absence of a regularization term.