PLDP-FL: Federated Learning with Personalized Local Differential Privacy
Entropy · 10.03.2023 · 25 (3) · 485
DOI: 10.3390/e25030485 ↗Kurz gefasst
Fachartikel in der Zeitschrift Entropy, online erschienen am 10.03.2023. Verfasst von Xiaoying Shen und 4 weiteren Autor:innen. Beteiligte Einrichtung: The State Key Laboratory of Integrated Service Networks, Xidian University, Xi'an 710071, China. Die Arbeit wurde bisher 57-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, National Natural Science Foundation of Chin, Open Research Fund of Key Laboratory of Cryptography of Zhejiang 553 Province.
Abstract (Originalsprache)
As a popular machine learning method, federated learning (FL) can effectively solve the issues of data silos and data privacy. However, traditional federated learning schemes cannot provide sufficient privacy protection. Furthermore, most secure federated learning schemes based on local differential privacy (LDP) ignore an important issue: they do not consider each client’s differentiated privacy requirements. This paper introduces a perturbation algorithm (PDPM) that satisfies personalized local differential privacy (PLDP), resolving the issue of inadequate or excessive privacy protection for some participants due to the same privacy budget set for all clients. The algorithm enables clients to adjust the privacy parameters according to the sensitivity of their data, thus allowing the scheme to provide personalized privacy protection. To ensure the privacy of the scheme, we have conducted a strict privacy proof and simulated the scheme on both synthetic and real data sets. Experiments have demonstrated that our scheme is successful in producing high-quality models and fulfilling the demands of personalized privacy protection.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- The State Key Laboratory of Integrated Service Networks, Xidian University, Xi'an 710071, China.
- The Key Laboratory of Cryptography of Zhejiang Province, Hangzhou Normal University, Hangzhou 310030, China.
- The School of Information Engineering, Xuchang University, Xuchang 461002, China.
- The Faculty of Intelligent Manufacturing, Wuyi University, Jiangmen 529000, China.
Förderung
- National Natural Science Foundation of China
- National Natural Science Foundation of Chin
- Open Research Fund of Key Laboratory of Cryptography of Zhejiang 553 Province
- Fundamental Research Funds for the Central Universities
- Open Research Fund of Key Laboratory of Cryptography of Zhejiang Province
Zitieren
APA
Shen, X., Jiang, H., Chen, Y., Wang, B., & Gao, L. (2023). PLDP-FL: Federated Learning with Personalized Local Differential Privacy. Entropy, 25(3), 485. https://doi.org/10.3390/e25030485
Vancouver
Shen X, Jiang H, Chen Y, Wang B, Gao L. PLDP-FL: Federated Learning with Personalized Local Differential Privacy. Entropy. 2023;25(3):485. doi:10.3390/e25030485
BibTeX
@article{Shen2023,
author = {Shen, Xiaoying and Jiang, Hang and Chen, Yange and Wang, Baocang and Gao, Le},
title = {PLDP-FL: Federated Learning with Personalized Local Differential Privacy},
journal = {Entropy},
year = {2023},
volume = {25},
number = {3},
pages = {485},
doi = {10.3390/e25030485}
}
EUVIMED