摘要
Federated learning (FL) is a framework which is used in distributed machine learning to obtain an optimal model from clients' local updates. As an efficient design in model convergence and data communication, cloud-edge-client hierarchical federated learning (HFL) attracts more attention than the typical cloud-client architecture. However, the HFL still poses threats to clients' sensitive data by analyzing the upload and download parameters. In this paper, to address information leakage effectively, we propose a novel privacy-preserving scheme based on the concept of differential pri-vacy (DP), adding Gaussian noises to the shared parameters when uploading them to edge and cloud servers and broadcasting them to clients. Our algorithm can obtain global differential privacy with adjustable noises in the architecture. We evaluate the performance on image classification tasks. In our experiment on the Modified National Institute of Standards and Technology (MNIST) dataset, we get 91% model accuracy. Compared to the previous two-layer HFL-DP, our design is more secure while as being accurate.
学科与关键词
- Adap DP-FL: Differentially Private Federated Learning with Adaptive NoiseFu, Jie; Chen, Zhili*; Han, Xiao21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications (IEEE TrustCom), Wuhan, PEOPLES R CHINA, 2022-12-09 To 2022-12-11.被引 1
- FL2DP: Privacy-Preserving Federated Learning Via Differential Privacy for Artificial IoTGu, Chen; Cui, Xuande; Zhu, Xiaoling; Hu, Donghui*IEEE Transactions on Industrial Informatics, 2023.被引 1
- A multifaceted survey on privacy preservation of federated learning: progress, challenges, and opportunitiesSaha, Sanchita; Hota, Ashlesha; Chattopadhyay, Arup Kumar; Nag, Amitava*; Nandi, SukumarArtificial Intelligence Review, 2024, 57(7): 184.被引 1
