No Cover Image

Journal article 111 views 33 downloads

APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios

Xueting Ma Orcid Logo, Guorui Ma, Yang Liu Orcid Logo, Shuhan Qi Orcid Logo

Entropy, Volume: 26, Issue: 8, Start page: 712

Swansea University Author: Yang Liu Orcid Logo

  • 67498.VoR.pdf

    PDF | Version of Record

    © 2024 by the authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.

    Download (20.75MB)

Check full text

DOI (Published version): 10.3390/e26080712

Abstract

With the rapid advancement of the Internet and big data technologies, traditional centralized machine learning methods are challenged when dealing with large-scale datasets. Federated Learning (FL), as an emerging distributed machine learning paradigm, enables multiple clients to collaboratively tra...

Full description

Published in: Entropy
ISSN: 1099-4300
Published: MDPI AG 2024
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa67498
first_indexed 2024-09-19T12:48:22Z
last_indexed 2024-11-25T14:20:17Z
id cronfa67498
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2024-09-19T13:48:51.6158179</datestamp><bib-version>v2</bib-version><id>67498</id><entry>2024-08-29</entry><title>APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios</title><swanseaauthors><author><sid>ba37dab58c9093dc63c79001565b75d4</sid><ORCID>0000-0003-2486-5765</ORCID><firstname>Yang</firstname><surname>Liu</surname><name>Yang Liu</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-08-29</date><deptcode>MACS</deptcode><abstract>With the rapid advancement of the Internet and big data technologies, traditional centralized machine learning methods are challenged when dealing with large-scale datasets. Federated Learning (FL), as an emerging distributed machine learning paradigm, enables multiple clients to collaboratively train a global model while preserving privacy. Edge computing, also recognized as a critical technology for handling massive datasets, has garnered significant attention. However, the heterogeneity of clients in edge computing environments can severely impact the performance of the resultant models. This study introduces an Adaptive Personalized Client-Selection and Model-Aggregation Algorithm, APCSMA, aimed at optimizing FL performance in edge computing settings. The algorithm evaluates clients&#x2019; contributions by calculating the real-time performance of local models and the cosine similarity between local and global models, and it designs a ContriFunc function to quantify each client&#x2019;s contribution. The server then selects clients and assigns weights during model aggregation based on these contributions. Moreover, the algorithm accommodates personalized needs in local model updates, rather than simply overwriting with the global model. Extensive experiments were conducted on the FashionMNIST and Cifar-10 datasets, simulating three data distributions with parameters dir = 0.1, 0.3, and 0.5. The accuracy improvements achieved were 3.9%, 1.9%, and 1.1% for the FashionMNIST dataset, and 31.9%, 8.4%, and 5.4% for the Cifar-10 dataset, respectively.</abstract><type>Journal Article</type><journal>Entropy</journal><volume>26</volume><journalNumber>8</journalNumber><paginationStart>712</paginationStart><paginationEnd/><publisher>MDPI AG</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>1099-4300</issnElectronic><keywords>edge computing; federated learning; client selection; model aggregation</keywords><publishedDay>21</publishedDay><publishedMonth>8</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-08-21</publishedDate><doi>10.3390/e26080712</doi><url/><notes/><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><funders>This research was funded by Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies (No. 2022B1212010005); Shenzhen Science and Technology Major Special Project (No. KJZD20230923114608017); National Natural Science Foundation of China (No. 62372139, No. 62376073); Natural Science Foundation of Guang-dong (No. 2024A1515030024); Shenzhen Stable Supporting Program (General Project) (No. GXWD20231130110352002); and Shenzhen Foundational Research Funding Under Grant (No. 20220818102414030, No. JCYJ20200109113427092).</funders><projectreference/><lastEdited>2024-09-19T13:48:51.6158179</lastEdited><Created>2024-08-29T16:45:01.5019490</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Xueting</firstname><surname>Ma</surname><orcid>0009-0003-1950-2908</orcid><order>1</order></author><author><firstname>Guorui</firstname><surname>Ma</surname><order>2</order></author><author><firstname>Yang</firstname><surname>Liu</surname><orcid>0000-0003-2486-5765</orcid><order>3</order></author><author><firstname>Shuhan</firstname><surname>Qi</surname><orcid>0000-0002-6903-145x</orcid><order>4</order></author></authors><documents><document><filename>67498__31387__d85322e9aa0b499f99005abaae111c3b.pdf</filename><originalFilename>67498.VoR.pdf</originalFilename><uploaded>2024-09-19T13:47:48.5278996</uploaded><type>Output</type><contentLength>21760228</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2024 by the authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2024-09-19T13:48:51.6158179 v2 67498 2024-08-29 APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios ba37dab58c9093dc63c79001565b75d4 0000-0003-2486-5765 Yang Liu Yang Liu true false 2024-08-29 MACS With the rapid advancement of the Internet and big data technologies, traditional centralized machine learning methods are challenged when dealing with large-scale datasets. Federated Learning (FL), as an emerging distributed machine learning paradigm, enables multiple clients to collaboratively train a global model while preserving privacy. Edge computing, also recognized as a critical technology for handling massive datasets, has garnered significant attention. However, the heterogeneity of clients in edge computing environments can severely impact the performance of the resultant models. This study introduces an Adaptive Personalized Client-Selection and Model-Aggregation Algorithm, APCSMA, aimed at optimizing FL performance in edge computing settings. The algorithm evaluates clients’ contributions by calculating the real-time performance of local models and the cosine similarity between local and global models, and it designs a ContriFunc function to quantify each client’s contribution. The server then selects clients and assigns weights during model aggregation based on these contributions. Moreover, the algorithm accommodates personalized needs in local model updates, rather than simply overwriting with the global model. Extensive experiments were conducted on the FashionMNIST and Cifar-10 datasets, simulating three data distributions with parameters dir = 0.1, 0.3, and 0.5. The accuracy improvements achieved were 3.9%, 1.9%, and 1.1% for the FashionMNIST dataset, and 31.9%, 8.4%, and 5.4% for the Cifar-10 dataset, respectively. Journal Article Entropy 26 8 712 MDPI AG 1099-4300 edge computing; federated learning; client selection; model aggregation 21 8 2024 2024-08-21 10.3390/e26080712 COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University Another institution paid the OA fee This research was funded by Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies (No. 2022B1212010005); Shenzhen Science and Technology Major Special Project (No. KJZD20230923114608017); National Natural Science Foundation of China (No. 62372139, No. 62376073); Natural Science Foundation of Guang-dong (No. 2024A1515030024); Shenzhen Stable Supporting Program (General Project) (No. GXWD20231130110352002); and Shenzhen Foundational Research Funding Under Grant (No. 20220818102414030, No. JCYJ20200109113427092). 2024-09-19T13:48:51.6158179 2024-08-29T16:45:01.5019490 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Xueting Ma 0009-0003-1950-2908 1 Guorui Ma 2 Yang Liu 0000-0003-2486-5765 3 Shuhan Qi 0000-0002-6903-145x 4 67498__31387__d85322e9aa0b499f99005abaae111c3b.pdf 67498.VoR.pdf 2024-09-19T13:47:48.5278996 Output 21760228 application/pdf Version of Record true © 2024 by the authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license. true eng https://creativecommons.org/licenses/by/4.0/
title APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
spellingShingle APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
Yang Liu
title_short APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
title_full APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
title_fullStr APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
title_full_unstemmed APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
title_sort APCSMA: Adaptive Personalized Client-Selection and Model-Aggregation Algorithm for Federated Learning in Edge Computing Scenarios
author_id_str_mv ba37dab58c9093dc63c79001565b75d4
author_id_fullname_str_mv ba37dab58c9093dc63c79001565b75d4_***_Yang Liu
author Yang Liu
author2 Xueting Ma
Guorui Ma
Yang Liu
Shuhan Qi
format Journal article
container_title Entropy
container_volume 26
container_issue 8
container_start_page 712
publishDate 2024
institution Swansea University
issn 1099-4300
doi_str_mv 10.3390/e26080712
publisher MDPI AG
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description With the rapid advancement of the Internet and big data technologies, traditional centralized machine learning methods are challenged when dealing with large-scale datasets. Federated Learning (FL), as an emerging distributed machine learning paradigm, enables multiple clients to collaboratively train a global model while preserving privacy. Edge computing, also recognized as a critical technology for handling massive datasets, has garnered significant attention. However, the heterogeneity of clients in edge computing environments can severely impact the performance of the resultant models. This study introduces an Adaptive Personalized Client-Selection and Model-Aggregation Algorithm, APCSMA, aimed at optimizing FL performance in edge computing settings. The algorithm evaluates clients’ contributions by calculating the real-time performance of local models and the cosine similarity between local and global models, and it designs a ContriFunc function to quantify each client’s contribution. The server then selects clients and assigns weights during model aggregation based on these contributions. Moreover, the algorithm accommodates personalized needs in local model updates, rather than simply overwriting with the global model. Extensive experiments were conducted on the FashionMNIST and Cifar-10 datasets, simulating three data distributions with parameters dir = 0.1, 0.3, and 0.5. The accuracy improvements achieved were 3.9%, 1.9%, and 1.1% for the FashionMNIST dataset, and 31.9%, 8.4%, and 5.4% for the Cifar-10 dataset, respectively.
published_date 2024-08-21T08:33:54Z
_version_ 1821393744601022464
score 11.071985

Similar Items