Are Equivariant Equilibrium Approximators Beneficial?
Abstract
Recently, remarkable progress has been made by approximating Nash equilibrium (NE), correlated equilibrium (CE), and coarse correlated equilibrium (CCE) through function approximation that trains a neural network to predict equilibria from game representations. Furthermore, equivariant architectures are widely adopted in designing such equilibrium approximators in normal-form games. In this paper, we theoretically characterize the benefits and limitations of equivariant equilibrium approximators. For the benefits, we show that they enjoy better generalizability than general ones and can achieve better approximations when the payoff distribution is permutation-invariant. For the limitations, we discuss their drawbacks in terms of equilibrium selection and social welfare. Together, our results help to understand the role of equivariance in equilibrium approximators.
Cite
Text
Duan et al. "Are Equivariant Equilibrium Approximators Beneficial?." International Conference on Machine Learning, 2023.Markdown
[Duan et al. "Are Equivariant Equilibrium Approximators Beneficial?." International Conference on Machine Learning, 2023.](https://mlanthology.org/icml/2023/duan2023icml-equivariant/)BibTeX
@inproceedings{duan2023icml-equivariant,
title = {{Are Equivariant Equilibrium Approximators Beneficial?}},
author = {Duan, Zhijian and Ma, Yunxuan and Deng, Xiaotie},
booktitle = {International Conference on Machine Learning},
year = {2023},
pages = {8747-8778},
volume = {202},
url = {https://mlanthology.org/icml/2023/duan2023icml-equivariant/}
}