Stochastic Extragradient with Random Reshuffling: Improved Convergence for Variational Inequalities

The Stochastic Extragradient (SEG) method is one of the most popular algorithms for solving finite-sum min-max optimization and variational inequality problems (VIPs) appearing in various machine learning tasks. However, existing convergence analyses of SEG focus on its with-replacement variants, wh...

Descripció completa

Dades bibliogràfiques

Autors principals:	Emmanouilidis, Konstantinos, Vidal, René, Loizou, Nicolas
Format:	Article
Publicat:	2024
Matèries:	Optimization and Control Computer Science and Game Theory Machine Learning
Accés en línia:	https://arxiv.org/abs/2403.07148 http://arxiv.org/abs/2403.07148
Etiquetes:	Afegir etiqueta Sense etiquetes, Sigues el primer a etiquetar aquest registre!