A Unified Approach to Reinforcement Learning, Quantal Response Equilibria, and Two-Player Zero-Sum GamesDownload PDF


22 Sept 2022, 12:35 (modified: 11 Nov 2022, 23:21)ICLR 2023 Conference Blind SubmissionReaders: Everyone
Keywords: reinforcement learning, quantal response equilibria, two-player zero-sum games, mirror descent, variational inequalities, Nash equilibria, algorithmic game theory, proximal gradient
TL;DR: A single algorithm for both single-agent reinforcement learning and approximating quantal response and Nash equilibria in two-player zero-sum games.
Abstract: Algorithms designed for single-agent reinforcement learning (RL) generally fail to converge to equilibria in two-player zero-sum (2p0s) games. On the other hand, game-theoretic algorithms for approximating Nash and regularized equilibria in 2p0s games are not typically competitive for RL and can be difficult to scale. As a result, algorithms for these two cases are generally developed and evaluated separately. In this work, we show that a single algorithm---a simple extension to mirror descent with proximal regularization that we call magnetic mirror descent (MMD)---can produce strong results in both settings, despite their fundamental differences. From a theoretical standpoint, we prove that MMD converges linearly to quantal response equilibria (i.e., entropy regularized Nash equilibria) in extensive-form games---this is the first time linear convergence has been proven for a first order solver. Moreover, applied as a tabular Nash equilibrium solver via self-play, we show empirically that MMD produces results competitive with CFR in both normal-form and extensive-form games---this is the first time that a standard RL algorithm has done so. Furthermore, for single-agent deep RL, on a small collection of Atari and Mujoco tasks, we show that MMD can produce results competitive with those of PPO. Lastly, for multi-agent deep RL, we show MMD can outperform NFSP in 3x3 Abrupt Dark Hex.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Submission Guidelines: Yes
Please Choose The Closest Area That Your Submission Falls Into: Theory (eg, control theory, learning theory, algorithmic game theory)
8 Replies