FedDefender: Client-Side Attack-Tolerant Federated Learning

TL;DR

FedDefender introduces a client-side defense mechanism for federated learning that enhances robustness against model poisoning attacks, especially when server-side defenses are insufficient or data distributions are non-i.i.d.

Contribution

It proposes a novel client-side defense method with attack-tolerant local updates and knowledge distillation, complementing server-side defenses to improve attack resilience.

Findings

Improves robustness against model poisoning attacks in federated learning.

Compatible with existing server-side defense strategies.

Effective across multiple datasets and real-world scenarios.

Abstract

Federated learning enables learning from decentralized data sources without compromising privacy, which makes it a crucial technique. However, it is vulnerable to model poisoning attacks, where malicious clients interfere with the training process. Previous defense mechanisms have focused on the server-side by using careful model aggregation, but this may not be effective when the data is not identically distributed or when attackers can access the information of benign clients. In this paper, we propose a new defense mechanism that focuses on the client-side, called FedDefender, to help benign clients train robust local models and avoid the adverse impact of malicious model updates from attackers, even when a server-side defense cannot identify or remove adversaries. Our method consists of two main components: (1) attack-tolerant local meta update and (2) attack-tolerant global knowledge distillation. These components are used to find noise-resilient model parameters while accurately extracting knowledge from a potentially corrupted global model. Our client-side defense strategy has a flexible structure and can work in conjunction with any existing server-side strategies. Evaluations of real-world scenarios across multiple datasets show that the proposed method enhances the robustness of federated learning against model poisoning attacks.