Efficient Privacy-Preserving Federated Learning With Improved Compressed Sensing
Publication Type
Journal Article
Publication Date
8-2023
Abstract
To solve the data silos issue in distributed machine learning with privacy leakage, privacy-preserving federated learning (PPFL) has been extensively explored in both academic and industrial fields. However, the existing PPFL solutions still suffer from high computation and communication overheads, which result in excessive consumption of communication bandwidth and slow down the training process of FL. To address these issues, we propose a secure and communication-efficient FL scheme using improved compressed sensing and CKKS homomorphic encryption. Specifically, we implement a lossy compression of the model by using discrete cosine transform, then use CKKS homomorphic encryption to encrypt the data transmitted between clients and center server due to its high efficiency and support for batch encryption. Formal security analysis proves that our scheme is secure against indistinguishability under chosen plaintext attack and extensive experiments demonstrate that our scheme achieves a high accuracy at 0.05% compression rate.
Keywords
CKKS, communication costs, compression sensing (CS), federated learning (FL), homomorphic encryption
Discipline
Artificial Intelligence and Robotics
Research Areas
Cybersecurity
Publication
IEEE Transactions on Industrial Informatics
ISSN
1551-3203
Identifier
10.1109/TII.2023.3297596
Publisher
Institute of Electrical and Electronics Engineers
Citation
ZHANG, Yifan.; MIAO, Yinbin.; LI, Xinghua.; WEI, Linfeng.; LIU, Zhiquan; CHOO, Kim-Kwang R.; and DENG, Robert H..
Efficient Privacy-Preserving Federated Learning With Improved Compressed Sensing. (2023). IEEE Transactions on Industrial Informatics.
Available at: https://ink.library.smu.edu.sg/sis_research/8291
Additional URL
https://doi.org/10.1109/TII.2023.3297596