Federated learning on non-IID and long-tailed data via dual-decoupling

IF 2.7 3区 工程技术 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS Frontiers of Information Technology & Electronic Engineering Pub Date : 2024-06-07 DOI:10.1631/fitee.2300284
Zhaohui Wang, Hongjiao Li, Jinguo Li, Renhao Hu, Baojin Wang
{"title":"Federated learning on non-IID and long-tailed data via dual-decoupling","authors":"Zhaohui Wang, Hongjiao Li, Jinguo Li, Renhao Hu, Baojin Wang","doi":"10.1631/fitee.2300284","DOIUrl":null,"url":null,"abstract":"<p>Federated learning (FL), a cutting-edge distributed machine learning training paradigm, aims to generate a global model by collaborating on the training of client models without revealing local private data. The cooccurrence of non-independent and identically distributed (non-IID) and long-tailed distribution in FL is one challenge that substantially degrades aggregate performance. In this paper, we present a corresponding solution called federated dual-decoupling via model and logit calibration (FedDDC) for non-IID and long-tailed distributions. The model is characterized by three aspects. First, we decouple the global model into the feature extractor and the classifier to fine-tune the components affected by the joint problem. For the biased feature extractor, we propose a client confidence re-weighting scheme to assist calibration, which assigns optimal weights to each client. For the biased classifier, we apply the classifier re-balancing method for fine-tuning. Then, we calibrate and integrate the client confidence re-weighted logits with the re-balanced logits to obtain the unbiased logits. Finally, we use decoupled knowledge distillation for the first time in the joint problem to enhance the accuracy of the global model by extracting the knowledge of the unbiased model. Numerous experiments demonstrate that on non-IID and long-tailed data in FL, our approach outperforms state-of-the-art methods.</p>","PeriodicalId":12608,"journal":{"name":"Frontiers of Information Technology & Electronic Engineering","volume":"52 1","pages":""},"PeriodicalIF":2.7000,"publicationDate":"2024-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers of Information Technology & Electronic Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1631/fitee.2300284","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 0

Abstract

Federated learning (FL), a cutting-edge distributed machine learning training paradigm, aims to generate a global model by collaborating on the training of client models without revealing local private data. The cooccurrence of non-independent and identically distributed (non-IID) and long-tailed distribution in FL is one challenge that substantially degrades aggregate performance. In this paper, we present a corresponding solution called federated dual-decoupling via model and logit calibration (FedDDC) for non-IID and long-tailed distributions. The model is characterized by three aspects. First, we decouple the global model into the feature extractor and the classifier to fine-tune the components affected by the joint problem. For the biased feature extractor, we propose a client confidence re-weighting scheme to assist calibration, which assigns optimal weights to each client. For the biased classifier, we apply the classifier re-balancing method for fine-tuning. Then, we calibrate and integrate the client confidence re-weighted logits with the re-balanced logits to obtain the unbiased logits. Finally, we use decoupled knowledge distillation for the first time in the joint problem to enhance the accuracy of the global model by extracting the knowledge of the unbiased model. Numerous experiments demonstrate that on non-IID and long-tailed data in FL, our approach outperforms state-of-the-art methods.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
通过双解耦对非 IID 和长尾数据进行联合学习
联合学习(FL)是一种前沿的分布式机器学习训练范式,旨在通过协作训练客户端模型来生成全局模型,而不会泄露本地私人数据。在联邦学习中,非独立同分布(non-IID)和长尾分布的共存是大幅降低总体性能的一个挑战。在本文中,我们针对非独立同分布和长尾分布提出了一种相应的解决方案,称为 "通过模型和对数校准进行联合双解耦"(FedDDC)。该模型有三个方面的特点。首先,我们将全局模型解耦为特征提取器和分类器,以微调受联合问题影响的部分。对于有偏差的特征提取器,我们提出了一种客户信心重新加权方案来帮助校准,该方案为每个客户分配了最佳权重。对于有偏差的分类器,我们采用分类器再平衡方法进行微调。然后,我们将客户信心再加权对数与再平衡对数进行校准和整合,以获得无偏对数。最后,我们首次在联合问题中使用解耦知识提炼法,通过提取无偏模型的知识来提高全局模型的准确性。大量实验证明,对于 FL 中的非 IID 和长尾数据,我们的方法优于最先进的方法。
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 去求助
来源期刊
Frontiers of Information Technology & Electronic Engineering
Frontiers of Information Technology & Electronic Engineering COMPUTER SCIENCE, INFORMATION SYSTEMSCOMPU-COMPUTER SCIENCE, SOFTWARE ENGINEERING
CiteScore
6.00
自引率
10.00%
发文量
1372
期刊介绍: Frontiers of Information Technology & Electronic Engineering (ISSN 2095-9184, monthly), formerly known as Journal of Zhejiang University SCIENCE C (Computers & Electronics) (2010-2014), is an international peer-reviewed journal launched by Chinese Academy of Engineering (CAE) and Zhejiang University, co-published by Springer & Zhejiang University Press. FITEE is aimed to publish the latest implementation of applications, principles, and algorithms in the broad area of Electrical and Electronic Engineering, including but not limited to Computer Science, Information Sciences, Control, Automation, Telecommunications. There are different types of articles for your choice, including research articles, review articles, science letters, perspective, new technical notes and methods, etc.
期刊最新文献
A novel overlapping minimization SMOTE algorithm for imbalanced classification A review on the developments and space applications of mid- and long-wavelength infrared detection technologies Detecting compromised accounts caused by phone number recycling on e-commerce platforms: taking Meituan as an example Flocking fragmentation formulation for a multi-robot system under multi-hop and lossy ad hoc networks Event-triggered distributed cross-dimensional formation control for heterogeneous multi-agent systems
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1