Escaping limit cycles: Global convergence for constrained nonconvex-nonconcave minimax problems

29 Sept 2021, 00:34 (modified: 16 Mar 2022, 13:47)ICLR 2022 SpotlightReaders: Everyone
Keywords: Minimax, Nonconvex-Nonconcave, Variational inequilities, Saddle point problem, First-order methods, Limit cycles
Abstract: This paper introduces a new extragradient-type algorithm for a class of nonconvex-nonconcave minimax problems. It is well-known that finding a local solution for general minimax problems is computationally intractable. This observation has recently motivated the study of structures sufficient for convergence of first order methods in the more general setting of variational inequalities when the so-called weak Minty variational inequality (MVI) holds. This problem class captures non-trivial structures as we demonstrate with examples, for which a large family of existing algorithms provably converge to limit cycles. Our results require a less restrictive parameter range in the weak MVI compared to what is previously known, thus extending the applicability of our scheme. The proposed algorithm is applicable to constrained and regularized problems, and involves an adaptive stepsize allowing for potentially larger stepsizes. Our scheme also converges globally even in settings where the underlying operator exhibits limit cycles.
One-sentence Summary: Under weak MVI we introduce a new extragradient-type algorithm that avoids limit cycles
Supplementary Material: zip
14 Replies