A federated learning scheme meets dynamic differential privacy
CAAI Transactions on Intelligence Technology · 2023-09 · 8 (3) · 1087-1100
DOI: 10.1049/cit2.12187 ↗Kurz gefasst
Fachartikel in der Zeitschrift CAAI Transactions on Intelligence Technology, online erschienen am 30.01.2023. Verfasst von Shengnan Guo und 5 weiteren Autor:innen. Die Arbeit wurde bisher 36-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Science and Technology Foundation of Guizhou Province, Guizhou Science and Technology Department.
Abstract (Originalsprache)
Federated learning is a widely used distributed learning approach in recent years, however, despite model training from collecting data become to gathering parameters, privacy violations may occur when publishing and sharing models. A dynamic approach is proposed to add Gaussian noise more effectively and apply differential privacy to federal deep learning. Concretely, it is abandoning the traditional way of equally distributing the privacy budget and adjusting the privacy budget to accommodate gradient descent federation learning dynamically, where the parameters depend on computation derived to avoid the impact on the algorithm that hyperparameters are created manually. It also incorporates adaptive threshold cropping to control the sensitivity, and finally, moments accountant is used to counting the consumed on the privacy‐preserving, and learning is stopped only if the by clients setting is reached, this allows the privacy budget to be adequately explored for model training. The experimental results on real datasets show that the method training has almost the same effect as the model learning of non‐privacy, which is significantly better than the differential privacy method used by TensorFlow.
Förderung
- National Natural Science Foundation of China
- Science and Technology Foundation of Guizhou Province
- Guizhou Science and Technology Department
Zitieren
APA
Guo, S., Wang, X., Long, S., Liu, H., Hai, L., & Sam, T. H. (2023). A federated learning scheme meets dynamic differential privacy. CAAI Transactions on Intelligence Technology, 8(3), 1087-1100. https://doi.org/10.1049/cit2.12187
Vancouver
Guo S, Wang X, Long S, Liu H, Hai L, Sam TH. A federated learning scheme meets dynamic differential privacy. CAAI Transactions on Intelligence Technology. 2023;8(3):1087-1100. doi:10.1049/cit2.12187
BibTeX
@article{Guo2023,
author = {Guo, Shengnan and Wang, Xibin and Long, Shigong and Liu, Hai and Hai, Liu and Sam, Toong Hai},
title = {A federated learning scheme meets dynamic differential privacy},
journal = {CAAI Transactions on Intelligence Technology},
year = {2023},
volume = {8},
number = {3},
pages = {1087--1100},
doi = {10.1049/cit2.12187}
}
EUVIMED