Variance Reduced Distributed Non-Convex Optimization Using Matrix Stepsizes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Hanmin, Karagulyan, Avetik, Richtárik, Peter
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909343644909568
author Li, Hanmin
Karagulyan, Avetik
Richtárik, Peter
author_facet Li, Hanmin
Karagulyan, Avetik
Richtárik, Peter
contents Matrix-stepsized gradient descent algorithms have been shown to have superior performance in non-convex optimization problems compared to their scalar counterparts. The det-CGD algorithm, as introduced by Li et al. (2023), leverages matrix stepsizes to perform compressed gradient descent for non-convex objectives and matrix-smooth problems in a federated manner. The authors establish the algorithm's convergence to a neighborhood of a weighted stationarity point under a convex condition for the symmetric and positive-definite matrix stepsize. In this paper, we propose two variance-reduced versions of the det-CGD algorithm, incorporating MARINA and DASHA methods. Notably, we establish theoretically and empirically, that det-MARINA and det-DASHA outperform MARINA, DASHA and the distributed det-CGD algorithms in terms of iteration and communication complexities.
format Preprint
id arxiv_https___arxiv_org_abs_2310_04614
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Variance Reduced Distributed Non-Convex Optimization Using Matrix Stepsizes
Li, Hanmin
Karagulyan, Avetik
Richtárik, Peter
Optimization and Control
90C26
Matrix-stepsized gradient descent algorithms have been shown to have superior performance in non-convex optimization problems compared to their scalar counterparts. The det-CGD algorithm, as introduced by Li et al. (2023), leverages matrix stepsizes to perform compressed gradient descent for non-convex objectives and matrix-smooth problems in a federated manner. The authors establish the algorithm's convergence to a neighborhood of a weighted stationarity point under a convex condition for the symmetric and positive-definite matrix stepsize. In this paper, we propose two variance-reduced versions of the det-CGD algorithm, incorporating MARINA and DASHA methods. Notably, we establish theoretically and empirically, that det-MARINA and det-DASHA outperform MARINA, DASHA and the distributed det-CGD algorithms in terms of iteration and communication complexities.
title Variance Reduced Distributed Non-Convex Optimization Using Matrix Stepsizes
topic Optimization and Control
90C26
url https://arxiv.org/abs/2310.04614