Optimality Conditions for Nonsmooth Nonconvex-Nonconcave Min-Max Problems and Generative Adversarial Networks

DOI10.1137/22M1482238zbMATH Open1522.90259arXiv2203.10914OpenAlexW4385721065MaRDI QIDQ6136233FDOQ6136233

Author name not available (Why is that?)

Publication date: 29 August 2023

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Abstract: This paper considers a class of nonsmooth nonconvex-nonconcave min-max problems in machine learning and games. We first provide sufficient conditions for the existence of global minimax points and local minimax points. Next, we establish the first-order and second-order optimality conditions for local minimax points by using directional derivatives. These conditions reduce to smooth min-max problems with Fr{'e}chet derivatives. We apply our theoretical results to generative adversarial networks (GANs) in which two neural networks contest with each other in a game. Examples are used to illustrate applications of the new theory for training GANs.

Full work available at URL: https://arxiv.org/abs/2203.10914

Recommendations

• Alternating Proximal-Gradient Steps for (Stochastic) Nonconvex-Concave Minimax Problems
• Weakly-convex-concave min-max optimization: provable algorithms and applications in machine learning
• The landscape of the proximal point method for nonconvex-nonconcave minimax optimization
• Two steps at a time-taking GAN training in stride with Tseng's method
• Efficient search of first-order Nash equilibria in nonconvex-concave smooth min-max problems

zbMATH Keywords

nonsmoothoptimality conditionmin-max problemgenerative adversarial networksnonconvex-nonconcave

Mathematics Subject Classification ID

Stochastic programming (90C15) Minimax problems in mathematical programming (90C47) Complementarity and equilibrium problems and variational inequalities (finite dimensions) (aspects of mathematical programming) (90C33) Numerical methods for variational inequalities and related problems (65K15)

Cites Work

• Variational Analysis
• Title not available (Why is that?)
• Prox-Method with Rate of Convergence O(1/t) for Variational Inequalities with Lipschitz Continuous Monotone Operators and Smooth Convex-Concave Saddle Point Problems
• Generalized Nash equilibrium problems
• Lectures on stochastic programming. Modeling and theory.
• A Generalized Second-Order Derivative in Nonsmooth Optimization
• Optimal Primal-Dual Methods for a Class of Saddle Point Problems
• Convergence Analysis of Sample Average Approximation of Two-Stage Stochastic Generalized Equations
• A study of piecewise linear-quadratic programs
• Exact Penalization of Generalized Nash Equilibrium Problems
• Hybrid Block Successive Approximation for One-Sided Non-Convex Min-Max Problems: Algorithms and Applications
• Bilevel optimization. Advances and next challenges
• On lower iteration complexity bounds for the convex concave saddle point problems
• On Gradient-Based Learning in Continuous Games
• Pure characteristics demand models and distributionally robust mathematical programs with stochastic complementarity constraints
• Weakly-convex–concave min–max optimization: provable algorithms and applications in machine learning
• Title not available (Why is that?)

Cited In (4)

• Data-driven distributionally robust multiproduct pricing problems under pure characteristics demand models
• Alternating Proximal-Gradient Steps for (Stochastic) Nonconvex-Concave Minimax Problems
• A quasi-Newton subspace trust region algorithm for nonmonotone variational inequalities in adversarial learning over box constraints
• Distributionally robust variational inequalities: relaxation, quantification and discretization

This page was built for publication: Optimality Conditions for Nonsmooth Nonconvex-Nonconcave Min-Max Problems and Generative Adversarial Networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136233)