Communication-Efficient Federated Optimization Over Semi-Decentralized Networks

IF 3 3区计算机科学 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

IEEE Transactions on Signal and Information Processing over Networks Pub Date : 2025-02-06 DOI:10.1109/TSIPN.2025.3539004

He Wang;Yuejie Chi

{"title":"Communication-Efficient Federated Optimization Over Semi-Decentralized Networks","authors":"He Wang;Yuejie Chi","doi":"10.1109/TSIPN.2025.3539004","DOIUrl":null,"url":null,"abstract":"In large-scale federated and decentralized learning, communication efficiency is one of the most challenging bottlenecks. While gossip communication—where agents can exchange information with their connected neighbors—is more cost-effective than communicating with the remote server, it often requires a greater number of communication rounds, especially for large and sparse networks. To tackle the trade-off, we examine the communication efficiency under a <italic>semi-decentralized</i> communication protocol, in which agents can perform both agent-to-agent and agent-to-server communication <italic>in a probabilistic manner</i>. We design a tailored communication-efficient algorithm over semi-decentralized networks, referred to as <monospace>PISCO</monospace>, which inherits the robustness to data heterogeneity thanks to gradient tracking and allows multiple local updates for saving communication. We establish the convergence rate of <monospace>PISCO</monospace> for nonconvex problems and show that <monospace>PISCO</monospace> enjoys a linear speedup in terms of the number of agents and local updates. Our numerical results highlight the superior communication efficiency of <monospace>PISCO</monospace> and its resilience to data heterogeneity and various network topologies.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"11 ","pages":"147-160"},"PeriodicalIF":3.0000,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Signal and Information Processing over Networks","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10877410/","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}

引用次数: 0

Abstract

In large-scale federated and decentralized learning, communication efficiency is one of the most challenging bottlenecks. While gossip communication—where agents can exchange information with their connected neighbors—is more cost-effective than communicating with the remote server, it often requires a greater number of communication rounds, especially for large and sparse networks. To tackle the trade-off, we examine the communication efficiency under a semi-decentralized communication protocol, in which agents can perform both agent-to-agent and agent-to-server communication in a probabilistic manner. We design a tailored communication-efficient algorithm over semi-decentralized networks, referred to as PISCO, which inherits the robustness to data heterogeneity thanks to gradient tracking and allows multiple local updates for saving communication. We establish the convergence rate of PISCO for nonconvex problems and show that PISCO enjoys a linear speedup in terms of the number of agents and local updates. Our numerical results highlight the superior communication efficiency of PISCO and its resilience to data heterogeneity and various network topologies.

查看原文本刊更多论文

半分散网络上的高效通信联邦优化

在大规模联合和分散学习中，沟通效率是最具挑战性的瓶颈之一。虽然八卦通信（代理可以与其连接的邻居交换信息）比与远程服务器通信更具成本效益，但它通常需要更多的通信轮数，特别是对于大型和稀疏的网络。为了解决这种权衡，我们研究了半分散通信协议下的通信效率，其中代理可以以概率方式执行代理到代理和代理到服务器的通信。我们在半分散网络上设计了一种定制的通信高效算法，称为PISCO，它继承了对数据异构的鲁棒性，这要归功于梯度跟踪，并允许多个本地更新以节省通信。我们建立了非凸问题的PISCO的收敛速度，并证明了PISCO在代理数量和局部更新方面具有线性加速。我们的数值结果显示了PISCO优越的通信效率以及它对数据异构和各种网络拓扑的弹性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE Transactions on Signal and Information Processing over Networks Computer Science-Computer Networks and Communications

CiteScore

5.80

自引率

12.50%

发文量

期刊介绍： The IEEE Transactions on Signal and Information Processing over Networks publishes high-quality papers that extend the classical notions of processing of signals defined over vector spaces (e.g. time and space) to processing of signals and information (data) defined over networks, potentially dynamically varying. In signal processing over networks, the topology of the network may define structural relationships in the data, or may constrain processing of the data. Topics include distributed algorithms for filtering, detection, estimation, adaptation and learning, model selection, data fusion, and diffusion or evolution of information over such networks, and applications of distributed signal processing.