A Descent-based method on the Duality Gap for solving zero-sum games

TL;DR

This paper introduces a descent-based algorithm leveraging the convexity of the duality gap for efficiently finding approximate equilibria in 2-player zero-sum games, with theoretical guarantees and promising experimental results.

Contribution

It proposes a novel steepest descent method on the duality gap for zero-sum games, improving convergence bounds and outperforming standard algorithms in large strategy spaces.

Findings

Achieves geometric decrease in duality gap

Outperforms OGDA in large strategy scenarios

Provides theoretical complexity bounds

Abstract

We focus on the design of algorithms for finding equilibria in 2-player zero-sum games. Although it is well known that such problems can be solved by a single linear program, there has been a surge of interest in recent years for simpler algorithms, motivated in part by applications in machine learning. Our work proposes such a method, inspired by the observation that the duality gap (a standard metric for evaluating convergence in min-max optimization problems) is a convex function for bilinear zero-sum games. To this end, we analyze a descent-based approach, variants of which have also been used as a subroutine in a series of algorithms for approximating Nash equilibria in general non-zero-sum games. In particular, we study a steepest descent approach, by finding the direction that minimises the directional derivative of the duality gap function. Our main theoretical result is that the derived algorithms achieve a geometric decrease in the duality gap and improved complexity bounds until we reach an approximate equilibrium. Finally, we complement this with an experimental evaluation, which provides promising findings. Our algorithm is comparable with (and in some cases outperforms) some of the standard approaches for solving 0-sum games, such as OGDA (Optimistic Gradient Descent/Ascent), even with thousands of available strategies per player.