FedGroup: Efficient Federated Learning via Decomposed Similarity-Based Clustering

Federated Learning (FL) enables the multiple participating devices to collaboratively contribute to a global neural network model while keeping the training data locally. Unlike the centralized training setting, the non-IID and imbalanced (statistical heterogeneity) training data of FL is distribute...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2021 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking (ISPA/BDCloud/SocialCom/SustainCom) s. 228 - 237
Hlavní autori: Duan, Moming, Liu, Duo, Ji, Xinyuan, Liu, Renping, Liang, Liang, Chen, Xianzhang, Tan, Yujuan
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 01.09.2021
Predmet:
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Federated Learning (FL) enables the multiple participating devices to collaboratively contribute to a global neural network model while keeping the training data locally. Unlike the centralized training setting, the non-IID and imbalanced (statistical heterogeneity) training data of FL is distributed in the federated network, which will increase the divergences between the local models and the global model, further degrading performance. In this paper, we propose a novel clustered federated learning (CFL) framework FedGroup, in which we 1) group the training of clients based on the similarities between the clients' optimization directions for high training performance; 2) construct a new data-driven distance measure to improve the efficiency of the client clustering procedure. 3) implement a newcomer device cold start mechanism based on the auxiliary global model for framework scalability and practicality.FedGroup can achieve improvements by dividing joint optimization into groups of sub-optimization and can be combined with FL optimizer FedProx. The convergence and complexity are analyzed to demonstrate the efficiency of our proposed framework. We also evaluate FedGroup and FedGrouProx (combined with FedProx) on several open datasets and made comparisons with related CFL frameworks. The results show that FedGroup can significantly improve absolute test accuracy by +14.1% on FEMNIST compared to FedAvg, +3.4% on Sentiment140 compared to FedProx, +6.9% on MNIST compared to FeSEM.
AbstractList Federated Learning (FL) enables the multiple participating devices to collaboratively contribute to a global neural network model while keeping the training data locally. Unlike the centralized training setting, the non-IID and imbalanced (statistical heterogeneity) training data of FL is distributed in the federated network, which will increase the divergences between the local models and the global model, further degrading performance. In this paper, we propose a novel clustered federated learning (CFL) framework FedGroup, in which we 1) group the training of clients based on the similarities between the clients' optimization directions for high training performance; 2) construct a new data-driven distance measure to improve the efficiency of the client clustering procedure. 3) implement a newcomer device cold start mechanism based on the auxiliary global model for framework scalability and practicality.FedGroup can achieve improvements by dividing joint optimization into groups of sub-optimization and can be combined with FL optimizer FedProx. The convergence and complexity are analyzed to demonstrate the efficiency of our proposed framework. We also evaluate FedGroup and FedGrouProx (combined with FedProx) on several open datasets and made comparisons with related CFL frameworks. The results show that FedGroup can significantly improve absolute test accuracy by +14.1% on FEMNIST compared to FedAvg, +3.4% on Sentiment140 compared to FedProx, +6.9% on MNIST compared to FeSEM.
Author Liang, Liang
Liu, Duo
Ji, Xinyuan
Duan, Moming
Chen, Xianzhang
Tan, Yujuan
Liu, Renping
Author_xml – sequence: 1
  givenname: Moming
  surname: Duan
  fullname: Duan, Moming
  organization: Chongqing University,College of Computer Science,Chongqing,China
– sequence: 2
  givenname: Duo
  surname: Liu
  fullname: Liu, Duo
  organization: Chongqing University,College of Computer Science,Chongqing,China
– sequence: 3
  givenname: Xinyuan
  surname: Ji
  fullname: Ji, Xinyuan
  organization: Xi'an Jiaotong University,School of Computer Science and Technology,Xi'an,China
– sequence: 4
  givenname: Renping
  surname: Liu
  fullname: Liu, Renping
  organization: Chongqing University,College of Computer Science,Chongqing,China
– sequence: 5
  givenname: Liang
  surname: Liang
  fullname: Liang, Liang
  organization: Chongqing University,College of Computer Science,Chongqing,China
– sequence: 6
  givenname: Xianzhang
  surname: Chen
  fullname: Chen, Xianzhang
  organization: Chongqing University,College of Computer Science,Chongqing,China
– sequence: 7
  givenname: Yujuan
  surname: Tan
  fullname: Tan, Yujuan
  organization: Chongqing University,College of Computer Science,Chongqing,China
BookMark eNotjctOhDAYhWuiCx3nCdzwAsXeoNTdDHNxEhI16HryF35ME6CkgMm8vZhxdU6-nMsDue19j4QYzmLOmXk-le8but3lrZ9rWvrKQZv7jpbzOIHrF5sIlvFYMMFjxpgSN2RtdMbTNFEy0Yrfk48D1sfg5-El2jeNqxz2U7QwDDBhHRUIoXf9d_TjINph5bvBjwsvXedaCG660C38gbxdTjEs0Udy10A74vpfV-TrsP_MX2nxdjzlm4I6weRElWYcrLGIElOWoKmFrJSSPEPgqBkIENpKYRPUBmsuRSWXjpVaC22UlSvydN11iHgegusgXM4mVUpnQv4CiLFXbw
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ISPA-BDCloud-SocialCom-SustainCom52081.2021.00042
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781665435741
1665435747
EndPage 237
ExternalDocumentID 9644782
Genre orig-research
GrantInformation_xml – fundername: National Natural Science Foundation of China
  funderid: 10.13039/501100001809
– fundername: Fundamental Research Funds for the Central Universities
  funderid: 10.13039/501100012226
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i203t-4701ab9bee3e605e9d23c44318ea1e70a2a27b32b5e79ed132c3470b3772794b3
IEDL.DBID RIE
ISICitedReferencesCount 78
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000766837400027&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 05:06:17 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-4701ab9bee3e605e9d23c44318ea1e70a2a27b32b5e79ed132c3470b3772794b3
PageCount 10
ParticipantIDs ieee_primary_9644782
PublicationCentury 2000
PublicationDate 2021-Sept.
PublicationDateYYYYMMDD 2021-09-01
PublicationDate_xml – month: 09
  year: 2021
  text: 2021-Sept.
PublicationDecade 2020
PublicationTitle 2021 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking (ISPA/BDCloud/SocialCom/SustainCom)
PublicationTitleAbbrev ISPA-BDCLOUD-SOCIALCOM-SUSTAINCOM
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
Score 2.2002928
Snippet Federated Learning (FL) enables the multiple participating devices to collaboratively contribute to a global neural network model while keeping the training...
SourceID ieee
SourceType Publisher
StartPage 228
SubjectTerms Collaborative work
Distributed Machine Learning
Federated Learning
Linear approximation
Neural networks
Performance evaluation
Scalability
Training
Training data
Title FedGroup: Efficient Federated Learning via Decomposed Similarity-Based Clustering
URI https://ieeexplore.ieee.org/document/9644782
WOSCitedRecordID wos000766837400027&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA5tEfGk0opv9uDR2N2k22y82ceil1KpQm8lj1lZqF1pu_39TrJLRfDiLQwbCLMJ33zJzDeE3JnQ8CwximYqipGghIoqbvoUpEG8j42JE-WbTYjJJJnP5bRB7ve1MADgk8_gwQ39W74tTOmuyroSwRsRrUmaQvSrWq1DImvZzO7LbPpEB6PhsigtrYpb8UzRWVWIhMOYIf4hJWSR1-hkvxqqeDxJj_-3khPS-SnMC6Z7yDklDVi1yWsK1l8hPQZjLwiBU4PUiURgHGmDWkH1I9jlKhiByyEvNmif5Z850lqMwulAOcNwWTrVBPy0Q97T8dvwmdadEmjOQr6lPRFGSksNwAH5CUjLuOlhbJCAikCEiikmNGc6BiHBIgM1HOdojrE1HkjNz0hrVazgnARJnInMQCQhhp4OIbE6inC7WSUUUidxQdrOHYuvSgxjUXvi8m_zFTly_q6Ssq5Ja7su4YYcmN0236xv_R_8BkkuoMs
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3fS8MwED7mFPVJZRN_2wcfjWuT1ba-uR9lwzkmm7C3kaRXKcxVtnV_v5e2TARffAtHA-Ga8N2X3H0HcKdtLWJfSxZLxyWCYksmhX5kGGjCe1dr15d5swlvOPSn02BUgfttLQwi5sln-GCG-Vt-lOrMXJU1AgJvQrQd2DWds8pqrX0ISuHMRn88ematTnueZhErylvpVLFxUYpEQ5cTAhIp5E6u0sl_tVTJESU8-t9ajqH-U5pnjbagcwIVXNTgLcQov0R6srq5JARNtUIjE0GRZGSVGqof1iaRVgdNFnm6Ivs4-UyI2FIczlrSGNrzzOgm0Kd1eA-7k3aPlb0SWMJtsWZNz3akChSiQGIoGERc6CZFBz5KBz1bcsk9Jbhy0QswIg6qBc1RgqJrOpJKnEJ1kS7wDCzfjb1YoxOgi01lox8px6ENF0lPEnnyzqFm3DH7KuQwZqUnLv4238JBb_I6mA36w5dLODS-L1K0rqC6XmZ4DXt6s05Wy5v8b34D1YukFA
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2021+IEEE+Intl+Conf+on+Parallel+%26+Distributed+Processing+with+Applications%2C+Big+Data+%26+Cloud+Computing%2C+Sustainable+Computing+%26+Communications%2C+Social+Computing+%26+Networking+%28ISPA%2FBDCloud%2FSocialCom%2FSustainCom%29&rft.atitle=FedGroup%3A+Efficient+Federated+Learning+via+Decomposed+Similarity-Based+Clustering&rft.au=Duan%2C+Moming&rft.au=Liu%2C+Duo&rft.au=Ji%2C+Xinyuan&rft.au=Liu%2C+Renping&rft.date=2021-09-01&rft.pub=IEEE&rft.spage=228&rft.epage=237&rft_id=info:doi/10.1109%2FISPA-BDCloud-SocialCom-SustainCom52081.2021.00042&rft.externalDocID=9644782