FedSlice: Protecting Federated Learning Models from Malicious Participants with Model Slicing

Abstract

Federated learning (FL) is a new crowdsourcing development paradigm for the DNN models, which is also called "software 2.0". In practice, the privacy of FL can be compromised by many attacks, such as free-rider attacks, adversarial attacks, gradient leakage attacks, and inference attacks. Conventional defensive techniques have low efficiency because they deploy heavy encryption techniques or rely on TEE. To improve the efficiency of protecting FL from the these attacks, this paper proposes FedSlice to prevent malicious participants from getting the whole server-side model while keeping the performance goal of FL. FedSlice breaks the server-side model into several slices and delivers one slice to each participant. Thus, a malicious participant can only get a subset of the server-side model, preventing them from effectively conducting effective attacks. We evaluate FedSlice against these attacks and results show that FedSlice provides effective defense: the server-side model leakage is reduced from 100% to 43.45%, the success rate of adversarial attacks is reduced from 100% to 11.66%, the average accuracy of membership inference is reduced from 71.91% to 51.58%, and the data leakage from shared gradients is reduced to the level of random guesses. Besides, FedSlice only introduces less than 2% accuracy loss and about 14% computation overhead. To the best of our knowledge, this is the first paper to discuss defense methods against these attacks to the FL framework.