Is Nash Equilibrium Approximator Learnable?
Computer Science and Game Theory
2023-03-15 v6 Machine Learning
Multiagent Systems
Abstract
In this paper, we investigate the learnability of the function approximator that approximates Nash equilibrium (NE) for games generated from a distribution. First, we offer a generalization bound using the Probably Approximately Correct (PAC) learning model. The bound describes the gap between the expected loss and empirical loss of the NE approximator. Afterward, we prove the agnostic PAC learnability of the Nash approximator. In addition to theoretical analysis, we demonstrate an application of NE approximator in experiments. The trained NE approximator can be used to warm-start and accelerate classical NE solvers. Together, our results show the practicability of approximating NE through function approximation.
Keywords
Cite
@article{arxiv.2108.07472,
title = {Is Nash Equilibrium Approximator Learnable?},
author = {Zhijian Duan and Wenhan Huang and Dinghuai Zhang and Yali Du and Jun Wang and Yaodong Yang and Xiaotie Deng},
journal= {arXiv preprint arXiv:2108.07472},
year = {2023}
}
Comments
Accepted by AAMAS 2023