Author
Anahtarcı, Berkay, Karıksız, Can Deha, Saldı, N.
Publication Date
2023-03
Publication Place
-
Springer
Subject
Discounted reward, Mean-field games, Q-learning, Regularized Markov decision processes
Type
Periodical
Language
English
Digital
Yes
Manuscript
No
Library
Özyeğin University
Library Asset ID
2153-0785
Record ID
60eb43e0-5e18-444f-9a58-62e50c99b500
Library Location
Natural and Mathematical Sciences
Date
2023-03
Notes
BAGEP Award of the Science Academy
Sample Text
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.
DOI
10.1007/s13235-022-00450-2
Cilt
13
