UAI 2021poster27 citations

Minimax sample complexity for turn-based stochastic game

Qiwen Cui, Lin F. Yang

Abstract

The empirical success of multi-agent reinforcement learning is encouraging, while few theoretical guarantees have been revealed. In this work, we prove that the plug-in solver approach, probably the most natural reinforcement learning algorithm, achieves minimax sample complexity for turn-based stochastic game (TBSG). Specifically, we perform planning in an empirical TBSG by utilizing a ‘simulator’ that allows sampling from arbitrary state-action pair. We show that the empirical Nash equilibrium strategy is an approximate Nash equilibrium strategy in the true TBSG and give both problem-dependent and problem-independent bound. We develop reward perturbation techniques to tackle the non-stationarity in the game and Taylor-expansion-type analysis to improve the dependence on approximation error. With these novel techniques, we prove the minimax sample complexity of turn-based stochastic game.

BibTeX
@InProceedings{pmlr-v161-cui21a,
  title = 	 {Minimax sample complexity for turn-based stochastic game},
  author =       {Cui, Qiwen and Yang, Lin F.},
  booktitle = 	 {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence},
  pages = 	 {1496--1504},
  year = 	 {2021},
  editor = 	 {de Campos, Cassio and Maathuis, Marloes H.},
  volume = 	 {161},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {27--30 Jul},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v161/cui21a/cui21a.pdf},
  url = 	 {https://proceedings.mlr.press/v161/cui21a.html},
  abstract = 	 {The empirical success of multi-agent reinforcement learning is encouraging, while few theoretical guarantees have been revealed. In this work, we prove that the plug-in solver approach, probably the most natural reinforcement learning algorithm, achieves minimax sample complexity for turn-based stochastic game (TBSG). Specifically, we perform planning in an empirical TBSG by utilizing a ‘simulator’ that allows sampling from arbitrary state-action pair. We show that the empirical Nash equilibrium strategy is an approximate Nash equilibrium strategy in the true TBSG and give both problem-dependent and problem-independent bound. We develop reward perturbation techniques to tackle the non-stationarity in the game and Taylor-expansion-type analysis to improve the dependence on approximation error. With these novel techniques, we prove the minimax sample complexity of turn-based stochastic game.}
}
Minimax sample complexity for turn-based stochastic game · UAI 2021