On the Linear Convergence of Extra-Gradient Methods for Nonconvex-Nonconcave Minimax Problems

TL;DR

This paper demonstrates that a damped version of the Extra-Gradient Method converges linearly for certain nonconvex-nonconcave minimax problems, offering a practical alternative to Proximal Point Methods.

Contribution

It establishes the linear convergence of a gradient-based Extra-Gradient Method for nonconvex-nonconcave minimax problems, expanding the theoretical understanding of first-order methods in this setting.

Findings

EGM converges linearly under specified conditions.

Avoids costly proximal computations.

Applicable to a broad class of minimax problems.

Abstract

Recently, minimax optimization received renewed focus due to modern applications in machine learning, robust optimization, and reinforcement learning. The scale of these applications naturally leads to the use of first-order methods. However, the nonconvexities and nonconcavities present in these problems, prevents the application of typical Gradient Descent-Ascent, which is known to diverge even in bilinear problems. Recently, it was shown that the Proximal Point Method (PPM) converges linearly for a family of nonconvex-nonconcave problems. In this paper, we study the convergence of a damped version of Extra-Gradient Method (EGM) which avoids potentially costly proximal computations, only relying on gradient evaluation. We show that EGM converges linearly for smooth minimax optimization problem satisfying the same nonconvex-nonconcave condition needed by PPM.