Federated learning via inexact ADMM
File(s)
Working Paper
Author(s)
Zhou, Shenglong
Li, Geoffrey Ye
Type
Working Paper
Abstract
One of the crucial issues in federated learning is how to develop efficient optimization algorithms. Most of the current ones require full device participation and/or impose strong assumptions for convergence. Different from the widely-used gradient descent-based algorithms, in this paper, we develop an inexact alternating direction method of multipliers (ADMM), which is both computation- and communication-efficient, capable of combating the stragglers' effect, and convergent under mild conditions. Furthermore, it has high numerical performance compared with several state-of-the-art algorithms for federated learning.
Date Issued
2023-02-07
Date Acceptance
2023-02-07
Citation
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023
ISSN
0162-8828
Publisher
ArXiv
Journal / Book Title
IEEE Transactions on Pattern Analysis and Machine Intelligence
Copyright Statement
©2022 The Author(s)
Identifier
http://arxiv.org/abs/2204.10607v1
Subjects
math.OC
math.OC
cs.LG
Publication Status
Published