MUSIC: Accelerated Convergence for Distributed Optimization With Inexact and Exact Methods

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Mou, Liao, Haibin, Ding, Zhengtao, Xiao, Yonggang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914702265679872
author Wu, Mou
Liao, Haibin
Ding, Zhengtao
Xiao, Yonggang
author_facet Wu, Mou
Liao, Haibin
Ding, Zhengtao
Xiao, Yonggang
contents Gradient-type distributed optimization methods have blossomed into one of the most important tools for solving a minimization learning task over a networked agent system. However, only one gradient update per iteration is difficult to achieve a substantive acceleration of convergence. In this paper, we propose an accelerated framework named as MUSIC allowing each agent to perform multiple local updates and a single combination in each iteration. More importantly, we equip inexact and exact distributed optimization methods into this framework, thereby developing two new algorithms that exhibit accelerated linear convergence and high communication efficiency. Our rigorous convergence analysis reveals the sources of steady-state errors arising from inexact policies and offers effective solutions. Numerical results based on synthetic and real datasets demonstrate both our theoretical motivations and analysis, as well as performance advantages.
format Preprint
id arxiv_https___arxiv_org_abs_2403_02589
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MUSIC: Accelerated Convergence for Distributed Optimization With Inexact and Exact Methods
Wu, Mou
Liao, Haibin
Ding, Zhengtao
Xiao, Yonggang
Optimization and Control
Artificial Intelligence
Gradient-type distributed optimization methods have blossomed into one of the most important tools for solving a minimization learning task over a networked agent system. However, only one gradient update per iteration is difficult to achieve a substantive acceleration of convergence. In this paper, we propose an accelerated framework named as MUSIC allowing each agent to perform multiple local updates and a single combination in each iteration. More importantly, we equip inexact and exact distributed optimization methods into this framework, thereby developing two new algorithms that exhibit accelerated linear convergence and high communication efficiency. Our rigorous convergence analysis reveals the sources of steady-state errors arising from inexact policies and offers effective solutions. Numerical results based on synthetic and real datasets demonstrate both our theoretical motivations and analysis, as well as performance advantages.
title MUSIC: Accelerated Convergence for Distributed Optimization With Inexact and Exact Methods
topic Optimization and Control
Artificial Intelligence
url https://arxiv.org/abs/2403.02589