Distributed Saddle-Point Problems: Lower Bounds, Near-Optimal and Robust Algorithms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Beznosikov, Aleksandr, Samokhin, Valentin, Gasnikov, Alexander
Format: Preprint
Published: 2020
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915257344065536
author Beznosikov, Aleksandr
Samokhin, Valentin
Gasnikov, Alexander
author_facet Beznosikov, Aleksandr
Samokhin, Valentin
Gasnikov, Alexander
contents This paper focuses on the distributed optimization of stochastic saddle point problems. The first part of the paper is devoted to lower bounds for the centralized and decentralized distributed methods for smooth (strongly) convex-(strongly) concave saddle point problems, as well as the near-optimal algorithms by which these bounds are achieved. Next, we present a new federated algorithm for centralized distributed saddle-point problems - Extra Step Local SGD. The theoretical analysis of the new method is carried out for strongly convex-strongly concave and non-convex-non-concave problems. In the experimental part of the paper, we show the effectiveness of our method in practice. In particular, we train GANs in a distributed manner.
format Preprint
id arxiv_https___arxiv_org_abs_2010_13112
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Distributed Saddle-Point Problems: Lower Bounds, Near-Optimal and Robust Algorithms
Beznosikov, Aleksandr
Samokhin, Valentin
Gasnikov, Alexander
Machine Learning
Distributed, Parallel, and Cluster Computing
Optimization and Control
This paper focuses on the distributed optimization of stochastic saddle point problems. The first part of the paper is devoted to lower bounds for the centralized and decentralized distributed methods for smooth (strongly) convex-(strongly) concave saddle point problems, as well as the near-optimal algorithms by which these bounds are achieved. Next, we present a new federated algorithm for centralized distributed saddle-point problems - Extra Step Local SGD. The theoretical analysis of the new method is carried out for strongly convex-strongly concave and non-convex-non-concave problems. In the experimental part of the paper, we show the effectiveness of our method in practice. In particular, we train GANs in a distributed manner.
title Distributed Saddle-Point Problems: Lower Bounds, Near-Optimal and Robust Algorithms
topic Machine Learning
Distributed, Parallel, and Cluster Computing
Optimization and Control
url https://arxiv.org/abs/2010.13112