Decentralized Federated Learning with Prototype Exchange
As AI applications become increasingly integrated into daily life, protecting user privacy while enabling collaborative model training has become a crucial challenge, especially in decentralized edge computing environments. Traditional federated learning (FL) approaches, which rely on centralized mo...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/2/237 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588069400739840 |
|---|---|
| author | Lu Qi Haoze Chen Hongliang Zou Shaohua Chen Xiaoying Zhang Hongyan Chen |
| author_facet | Lu Qi Haoze Chen Hongliang Zou Shaohua Chen Xiaoying Zhang Hongyan Chen |
| author_sort | Lu Qi |
| collection | DOAJ |
| description | As AI applications become increasingly integrated into daily life, protecting user privacy while enabling collaborative model training has become a crucial challenge, especially in decentralized edge computing environments. Traditional federated learning (FL) approaches, which rely on centralized model aggregation, struggle in such settings due to bandwidth limitations, data heterogeneity, and varying device capabilities among edge nodes. To address these issues, we propose PearFL, a decentralized FL framework that enhances collaboration and model generalization by introducing prototype exchange mechanisms. PearFL allows each client to share lightweight prototype information with its neighbors, minimizing communication overhead and improving model consistency across distributed devices. Experimental evaluations on benchmark datasets, including MNIST, CIFAR-10, and CIFAR-100, demonstrate that PearFL achieves superior communication efficiency, convergence speed, and accuracy compared to conventional FL methods. These results confirm PearFL’s efficacy as a scalable solution for decentralized learning in heterogeneous and resource-constrained environments. |
| format | Article |
| id | doaj-art-81f37d702b8345a48282fbfb41d623f0 |
| institution | Kabale University |
| issn | 2227-7390 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-81f37d702b8345a48282fbfb41d623f02025-01-24T13:39:51ZengMDPI AGMathematics2227-73902025-01-0113223710.3390/math13020237Decentralized Federated Learning with Prototype ExchangeLu Qi0Haoze Chen1Hongliang Zou2Shaohua Chen3Xiaoying Zhang4Hongyan Chen5College of Modern Science and Technology, China Jiliang University, Yiwu 322002, ChinaCollege of Modern Science and Technology, China Jiliang University, Yiwu 322002, ChinaCollege of Modern Science and Technology, China Jiliang University, Yiwu 322002, ChinaCollege of Modern Science and Technology, China Jiliang University, Yiwu 322002, ChinaCollege of Modern Science and Technology, China Jiliang University, Yiwu 322002, ChinaCollege of Modern Science and Technology, China Jiliang University, Yiwu 322002, ChinaAs AI applications become increasingly integrated into daily life, protecting user privacy while enabling collaborative model training has become a crucial challenge, especially in decentralized edge computing environments. Traditional federated learning (FL) approaches, which rely on centralized model aggregation, struggle in such settings due to bandwidth limitations, data heterogeneity, and varying device capabilities among edge nodes. To address these issues, we propose PearFL, a decentralized FL framework that enhances collaboration and model generalization by introducing prototype exchange mechanisms. PearFL allows each client to share lightweight prototype information with its neighbors, minimizing communication overhead and improving model consistency across distributed devices. Experimental evaluations on benchmark datasets, including MNIST, CIFAR-10, and CIFAR-100, demonstrate that PearFL achieves superior communication efficiency, convergence speed, and accuracy compared to conventional FL methods. These results confirm PearFL’s efficacy as a scalable solution for decentralized learning in heterogeneous and resource-constrained environments.https://www.mdpi.com/2227-7390/13/2/237federated learningdistributed machine learningprototype exchange |
| spellingShingle | Lu Qi Haoze Chen Hongliang Zou Shaohua Chen Xiaoying Zhang Hongyan Chen Decentralized Federated Learning with Prototype Exchange Mathematics federated learning distributed machine learning prototype exchange |
| title | Decentralized Federated Learning with Prototype Exchange |
| title_full | Decentralized Federated Learning with Prototype Exchange |
| title_fullStr | Decentralized Federated Learning with Prototype Exchange |
| title_full_unstemmed | Decentralized Federated Learning with Prototype Exchange |
| title_short | Decentralized Federated Learning with Prototype Exchange |
| title_sort | decentralized federated learning with prototype exchange |
| topic | federated learning distributed machine learning prototype exchange |
| url | https://www.mdpi.com/2227-7390/13/2/237 |
| work_keys_str_mv | AT luqi decentralizedfederatedlearningwithprototypeexchange AT haozechen decentralizedfederatedlearningwithprototypeexchange AT hongliangzou decentralizedfederatedlearningwithprototypeexchange AT shaohuachen decentralizedfederatedlearningwithprototypeexchange AT xiaoyingzhang decentralizedfederatedlearningwithprototypeexchange AT hongyanchen decentralizedfederatedlearningwithprototypeexchange |