PEPFL:A framework for a practical and efficient privacy-preserving federated learning

Yange Chen ¹Baocang Wang ²Hang Jiang ³Pu Duan ⁴Yuan Ping ⁵Zhiyong Hong⁶

扫码查看

作者信息

1. State Key Laboratory of Integrated Service Networks,Xidian University,Xi'an,710071,China;School of Information Engineering,Xuchang University,Xuchang,461000,China;School of Telecommunications Engineering,Xidian University,Xi'an,710071,China
2. State Key Laboratory of Integrated Service Networks,Xidian University,Xi'an,710071,China;School of Information Engineering,Xuchang University,Xuchang,461000,China
3. School of Telecommunications Engineering,Xidian University,Xi'an,710071,China
4. Secure Collaborative Intelligence Laboratory,Ant Group,Hangzhou,310000,China
5. School of Information Engineering,Xuchang University,Xuchang,461000,China
6. Facility of Intelligence Manufacture Wuyi University,Jiangmen,529020,China
折叠

Abstract

As an emerging joint learning model,federated learning is a promising way to combine model parameters of different users for training and inference without collecting users'original data.However,a practical and efficient solution has not been established in previous work due to the absence of efficient matrix computation and cryptography schemes in the privacy-preserving federated learning model,especially in partially homomorphic cryptosystems.In this paper,we propose a Practical and Efficient Privacy-preserving Federated Learning(PEPFL)framework.First,we present a lifted distributed ElGamal cryptosystem for federated learning,which can solve the multi-key problem in federated learning.Secondly,we develop a Practical Partially Single Instruction Multiple Data(PSIMD)parallelism scheme that can encode a plaintext matrix into single plaintext for encryption,improving the encryption efficiency and reducing the communication cost in partially homomorphic cryptosys-tem.In addition,based on the Convolutional Neural Network(CNN)and the designed cryptosystem,a novel privacy-preserving federated learning framework is designed by using Momentum Gradient Descent(MGD).Finally,we evaluate the security and performance of PEPFL.The experiment results demonstrate that the scheme is practicable,effective,and secure with low communication and computation costs.

Key words

Federated learning/Partially single instruction multiple data/Momentum gradient descent/ElGamal/Multi-key/Homomorphic encryption

引用本文复制引用

基金项目

National Natural Science Foundation of China(U19B2021)

Key Research and Development Program of Shaanxi(2020ZDLGY08-04)

Key Technologies R & D Program of He'nan Province(212102210084)

Innovation Scientists and Technicians Troop Construction Projects of Henan Province()

出版年

2024

数字通信与网络(英文)

ISSN：

参考文献量53

段落导航