Privacy-Preserving Federated Compressed Learning Against Data Reconstruction Attacks Based on Secure Data

Federated learning is a new distributed learning framework with data privacy preserving in which multiple users collaboratively train models without sharing data. However, recent studies highlight potential privacy leakage through shared gradient information. Several defense strategies, including gradient information encryption and perturbation, have been suggested. But these strategies either involve high complexity or are susceptible to attacks. To counter these challenges, we propose to train on secure compressive measurements by compressed learning, thereby achieving local data privacy protection with slight performance degradation. A feasible method to boost performance in compressed learning is the joint optimization of the sampling matrix and the inference network during the training phase, but this may suffer from data reconstruction attacks again. Thus, we further incorporate a traditional lightweight encryption scheme to protect data privacy. Experiments conducted on MNIST and FMNIST datasets substantiate that our schemes achieve a satisfactory balance between privacy protection and model performance.