CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging
The personalized federated averaging algorithm integrates a federated averaging approach with a model-agnostic meta-learning technique. In real-world heterogeneous scenarios, it is essential to implement additional privacy protection techniques for personalized federated learning. We propose a novel...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-11-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/22/3630 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850227130250559488 |
|---|---|
| author | Wenquan Shen Shuhui Wu Yuanhong Tao |
| author_facet | Wenquan Shen Shuhui Wu Yuanhong Tao |
| author_sort | Wenquan Shen |
| collection | DOAJ |
| description | The personalized federated averaging algorithm integrates a federated averaging approach with a model-agnostic meta-learning technique. In real-world heterogeneous scenarios, it is essential to implement additional privacy protection techniques for personalized federated learning. We propose a novel differentially private federated meta-learning scheme, CLDP-pFedAvg, which achieves client-level differential privacy guarantees for federated learning involving large heterogeneous clients. The client-level differentially private meta-based FedAvg algorithm enables clients to upload local model parameters for aggregation securely. Furthermore, we provide a convergence analysis of the clipping-enabled differentially private meta-based FedAvg algorithm. The proposed strategy is evaluated across various datasets, and the findings indicate that our approach offers improved privacy protection while maintaining model accuracy. |
| format | Article |
| id | doaj-art-9f8a3ff72db94e51bc312c19d72c55ce |
| institution | OA Journals |
| issn | 2227-7390 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-9f8a3ff72db94e51bc312c19d72c55ce2025-08-20T02:04:54ZengMDPI AGMathematics2227-73902024-11-011222363010.3390/math12223630CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated AveragingWenquan Shen0Shuhui Wu1Yuanhong Tao2School of Science, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaSchool of Science, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaSchool of Science, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaThe personalized federated averaging algorithm integrates a federated averaging approach with a model-agnostic meta-learning technique. In real-world heterogeneous scenarios, it is essential to implement additional privacy protection techniques for personalized federated learning. We propose a novel differentially private federated meta-learning scheme, CLDP-pFedAvg, which achieves client-level differential privacy guarantees for federated learning involving large heterogeneous clients. The client-level differentially private meta-based FedAvg algorithm enables clients to upload local model parameters for aggregation securely. Furthermore, we provide a convergence analysis of the clipping-enabled differentially private meta-based FedAvg algorithm. The proposed strategy is evaluated across various datasets, and the findings indicate that our approach offers improved privacy protection while maintaining model accuracy.https://www.mdpi.com/2227-7390/12/22/3630personalized federated averagingclient-level differential privacymeta-learning |
| spellingShingle | Wenquan Shen Shuhui Wu Yuanhong Tao CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging Mathematics personalized federated averaging client-level differential privacy meta-learning |
| title | CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging |
| title_full | CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging |
| title_fullStr | CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging |
| title_full_unstemmed | CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging |
| title_short | CLDP-pFedAvg: Safeguarding Client Data Privacy in Personalized Federated Averaging |
| title_sort | cldp pfedavg safeguarding client data privacy in personalized federated averaging |
| topic | personalized federated averaging client-level differential privacy meta-learning |
| url | https://www.mdpi.com/2227-7390/12/22/3630 |
| work_keys_str_mv | AT wenquanshen cldppfedavgsafeguardingclientdataprivacyinpersonalizedfederatedaveraging AT shuhuiwu cldppfedavgsafeguardingclientdataprivacyinpersonalizedfederatedaveraging AT yuanhongtao cldppfedavgsafeguardingclientdataprivacyinpersonalizedfederatedaveraging |