Adaptive clustering federated learning via similarity acceleration
In order to solve the problem of model performance degradation caused by data heterogeneity in the federated learning process, it is necessary to consider personalizing in the federated model.A new adaptively clustering federated learning (ACFL) algorithm via similarity acceleration was proposed, ac...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2024-03-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024069/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841540052624080896 |
|---|---|
| author | Suxia ZHU Binke GU Guanglu SUN |
| author_facet | Suxia ZHU Binke GU Guanglu SUN |
| author_sort | Suxia ZHU |
| collection | DOAJ |
| description | In order to solve the problem of model performance degradation caused by data heterogeneity in the federated learning process, it is necessary to consider personalizing in the federated model.A new adaptively clustering federated learning (ACFL) algorithm via similarity acceleration was proposed, achieving adaptive acceleration clustering based on geometric properties of local updates and the positive feedback mechanism during clients federated training.By dividing clients into different task clusters, clients with similar data distribution in the same cluster was cooperated to improve the performance of federated model.It did not need to determine the number of clusters in advance and iteratively divide the clients, so as to avoid the problems of high computational cost and slow convergence speed in the existing clustering federation methods while ensuring the performance of models.The effectiveness of ACFL was verified by using deep convolutional neural networks on commonly used datasets.The results show that the performance of ACFL is comparable to the clustered federated learning (CFL) algorithm, it is better than the traditional iterative federated cluster algorithm (IFCA), and has faster convergence speed. |
| format | Article |
| id | doaj-art-7c7491b5540a47b29d0d77425d8ff917 |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2024-03-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-7c7491b5540a47b29d0d77425d8ff9172025-01-14T06:21:57ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2024-03-014519720759296798Adaptive clustering federated learning via similarity accelerationSuxia ZHUBinke GUGuanglu SUNIn order to solve the problem of model performance degradation caused by data heterogeneity in the federated learning process, it is necessary to consider personalizing in the federated model.A new adaptively clustering federated learning (ACFL) algorithm via similarity acceleration was proposed, achieving adaptive acceleration clustering based on geometric properties of local updates and the positive feedback mechanism during clients federated training.By dividing clients into different task clusters, clients with similar data distribution in the same cluster was cooperated to improve the performance of federated model.It did not need to determine the number of clusters in advance and iteratively divide the clients, so as to avoid the problems of high computational cost and slow convergence speed in the existing clustering federation methods while ensuring the performance of models.The effectiveness of ACFL was verified by using deep convolutional neural networks on commonly used datasets.The results show that the performance of ACFL is comparable to the clustered federated learning (CFL) algorithm, it is better than the traditional iterative federated cluster algorithm (IFCA), and has faster convergence speed.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024069/federated learningpersonalizationclusteringgeometric characteristicpositive feedback |
| spellingShingle | Suxia ZHU Binke GU Guanglu SUN Adaptive clustering federated learning via similarity acceleration Tongxin xuebao federated learning personalization clustering geometric characteristic positive feedback |
| title | Adaptive clustering federated learning via similarity acceleration |
| title_full | Adaptive clustering federated learning via similarity acceleration |
| title_fullStr | Adaptive clustering federated learning via similarity acceleration |
| title_full_unstemmed | Adaptive clustering federated learning via similarity acceleration |
| title_short | Adaptive clustering federated learning via similarity acceleration |
| title_sort | adaptive clustering federated learning via similarity acceleration |
| topic | federated learning personalization clustering geometric characteristic positive feedback |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024069/ |
| work_keys_str_mv | AT suxiazhu adaptiveclusteringfederatedlearningviasimilarityacceleration AT binkegu adaptiveclusteringfederatedlearningviasimilarityacceleration AT guanglusun adaptiveclusteringfederatedlearningviasimilarityacceleration |