ICML 2022spotlight25 citations
A Self-Play Posterior Sampling Algorithm for Zero-Sum Markov Games
Wei Xiong, Han Zhong, Chengshuai Shi, Cong Shen, Tong Zhang
Abstract
Existing studies on provably efficient algorithms for Markov games (MGs) almost exclusively build on the “optimism in the face of uncertainty” (OFU) principle. This work focuses on a distinct approach of posterior sampling, which is celebrated in many bandits and reinforcement learning settings but remains under-explored for MGs. Specifically, for episodic two-player zero-sum MGs, a novel posterior sampling algorithm is developed with
BibTeX
@InProceedings{pmlr-v162-xiong22b,
title = {A Self-Play Posterior Sampling Algorithm for Zero-Sum {M}arkov Games},
author = {Xiong, Wei and Zhong, Han and Shi, Chengshuai and Shen, Cong and Zhang, Tong},
booktitle = {Proceedings of the 39th International Conference on Machine Learning},
pages = {24496--24523},
year = {2022},
editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
volume = {162},
series = {Proceedings of Machine Learning Research},
month = {17--23 Jul},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v162/xiong22b/xiong22b.pdf},
url = {https://proceedings.mlr.press/v162/xiong22b.html},
abstract = {Existing studies on provably efficient algorithms for Markov games (MGs) almost exclusively build on the “optimism in the face of uncertainty” (OFU) principle. This work focuses on a distinct approach of posterior sampling, which is celebrated in many bandits and reinforcement learning settings but remains under-explored for MGs. Specifically, for episodic two-player zero-sum MGs, a novel posterior sampling algorithm is developed with