Equilibrium Finding in Normal-Form Games Via Greedy Regret Minimization

TL;DR

This paper introduces a greedy regret minimization approach for finding equilibria in normal-form games, maintaining convergence guarantees and significantly improving empirical performance on large and complex games.

Contribution

It extends regret minimization with a greedy weighting scheme that enhances practical efficiency while preserving theoretical convergence guarantees.

Findings

Greedy weighting outperforms previous methods with sampling.

Significant empirical improvements on large random and Diplomacy subgames.

Method retains all existing convergence rate guarantees.

Abstract

We extend the classic regret minimization framework for approximating equilibria in normal-form games by greedily weighing iterates based on regrets observed at runtime. Theoretically, our method retains all previous convergence rate guarantees. Empirically, experiments on large randomly generated games and normal-form subgames of the AI benchmark Diplomacy show that greedy weights outperforms previous methods whenever sampling is used, sometimes by several orders of magnitude.