Gradient-Leakage Resilient Federated Learning

Wenqi Wei, Ling Liu, Yanzhao Wu, Gong Su, Arun Iyengar

2021 · 100 citations · 34 references

Concepts

TL;DR

Federated learning keeps data on devices and shares only local updates, yet recent studies show gradient leakages can still compromise client privacy. This work introduces Fed‑CDP, a per‑example client differential privacy method for federated learning designed to resist gradient‑leakage attacks. The authors identify three leakage threat types, propose Fed‑CDP with a dynamic decay noise‑injection policy, formally analyze its privacy‑utility trade‑off, and compare it to the conventional Fed‑SDP scheme. Experiments on five benchmark datasets demonstrate that Fed‑CDP outperforms Fed‑SDP in gradient‑leakage resilience while maintaining competitive accuracy.

Abstract

Federated learning(FL) is an emerging distributed learning paradigm with default client privacy because clients can keep sensitive data on their devices and only share local training parameter updates with the federated server. However, recent studies reveal that gradient leakages in FL may compromise the privacy of client training data. This paper presents a gradient leakage resilient approach to privacy-preserving federated learning with per training example-based client differential privacy, coined as Fed-CDP. It makes three original contributions. First, we identify three types of client gradient leakage threats in federated learning even with encrypted client-server communications. We articulate when and why the conventional server coordinated differential privacy approach, coined as Fed-SDP, is insufficient to protect the privacy of the training data. Second, we introduce Fed-CDP, the per example-based client differential privacy algorithm, and provide a formal analysis of Fed-CDP with the (∊,δ) differential privacy guarantee, and a formal comparison between Fed-CDP and Fed-SDP in terms of privacy accounting. Third, we formally analyze the privacy-utility tradeoff for providing differential privacy guarantee by Fed-CDP and present a dynamic decay noise-injection policy to further improve the accuracy and resiliency of Fed-CDP. We evaluate and compare Fed-CDP and Fed-CDP(decay) with Fed-SDP in terms of differential privacy guarantee and gradient leakage resilience over five benchmark datasets. The results show that the Fed-CDP approach outperforms conventional Fed-SDP in terms of resilience to client gradient leakages while offering competitive accuracy performance in federated learning.

References

34