Information Entropy-Based Scheduling for Communication-Efficient Decentralized Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nagar, Jaiprakash, Chen, Zheng, Kountouris, Marios, Stavrou, Photios A.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911154742231040
author Nagar, Jaiprakash
Chen, Zheng
Kountouris, Marios
Stavrou, Photios A.
author_facet Nagar, Jaiprakash
Chen, Zheng
Kountouris, Marios
Stavrou, Photios A.
contents This paper addresses decentralized stochastic gradient descent (D-SGD) over resource-constrained networks by introducing node-based and link-based scheduling strategies to enhance communication efficiency. In each iteration of the D-SGD algorithm, only a few disjoint subsets of nodes or links are randomly activated, subject to a given communication cost constraint. We propose a novel importance metric based on information entropy to determine node and link scheduling probabilities. We validate the effectiveness of our approach through extensive simulations, comparing it against state-of-the-art methods, including betweenness centrality (BC) for node scheduling and \textit{MATCHA} for link scheduling. The results show that our method consistently outperforms the BC-based method in the node scheduling case, achieving faster convergence with up to 60\% lower communication budgets. At higher communication budgets (above 60\%), our method maintains comparable or superior performance. In the link scheduling case, our method delivers results that are superior to or on par with those of \textit{MATCHA}.
format Preprint
id arxiv_https___arxiv_org_abs_2507_17426
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Information Entropy-Based Scheduling for Communication-Efficient Decentralized Learning
Nagar, Jaiprakash
Chen, Zheng
Kountouris, Marios
Stavrou, Photios A.
Information Theory
Machine Learning
Networking and Internet Architecture
This paper addresses decentralized stochastic gradient descent (D-SGD) over resource-constrained networks by introducing node-based and link-based scheduling strategies to enhance communication efficiency. In each iteration of the D-SGD algorithm, only a few disjoint subsets of nodes or links are randomly activated, subject to a given communication cost constraint. We propose a novel importance metric based on information entropy to determine node and link scheduling probabilities. We validate the effectiveness of our approach through extensive simulations, comparing it against state-of-the-art methods, including betweenness centrality (BC) for node scheduling and \textit{MATCHA} for link scheduling. The results show that our method consistently outperforms the BC-based method in the node scheduling case, achieving faster convergence with up to 60\% lower communication budgets. At higher communication budgets (above 60\%), our method maintains comparable or superior performance. In the link scheduling case, our method delivers results that are superior to or on par with those of \textit{MATCHA}.
title Information Entropy-Based Scheduling for Communication-Efficient Decentralized Learning
topic Information Theory
Machine Learning
Networking and Internet Architecture
url https://arxiv.org/abs/2507.17426