Accelerated variance-reduced methods for saddle-point problems

Accelerated variance-reduced methods for saddle-point problems

We consider composite minimax optimization problems where the goal is to find a saddle-point of a large sum of non-bilinear objective functions augmented by simple composite regularizers for the primal and dual variables. For such problems, under the average-smoothness assumption, we propose acceler...

Full description

Bibliographic Details
Main Authors: Ekaterina Borodich, Vladislav Tominin, Yaroslav Tominin, Dmitry Kovalev, Alexander Gasnikov, Pavel Dvurechensky
Format: Article
Language:English
Published: Elsevier 2022-01-01
Series:EURO Journal on Computational Optimization
Subjects:
Saddle-point problem
Minimax optimization
Composite optimization
Stochastic variance-reduced algorithms
Accelerated algorithms
Online Access:http://www.sciencedirect.com/science/article/pii/S2192440622000247

Internet

http://www.sciencedirect.com/science/article/pii/S2192440622000247

Similar Items