Fedward: Flexible Federated Backdoor Defense Framework with Non-IID Data

TL;DR

Fedward is a flexible federated learning defense framework that effectively eliminates backdoors in non-IID data scenarios by decomposing attacks and employing adaptive defense mechanisms, significantly improving robustness against backdoor attacks.

Contribution

The paper introduces Fedward, a novel framework that decomposes federated backdoor attacks and employs adaptive techniques to eliminate backdoors in non-IID data settings.

Findings

Achieves 96.98% success rate reduction on MNIST

Improves clustering defense effectiveness by 33-75%

Maintains performance in Non-IID scenarios

Abstract

Federated learning (FL) enables multiple clients to collaboratively train deep learning models while considering sensitive local datasets' privacy. However, adversaries can manipulate datasets and upload models by injecting triggers for federated backdoor attacks (FBA). Existing defense strategies against FBA consider specific and limited attacker models, and a sufficient amount of noise to be injected only mitigates rather than eliminates FBA. To address these deficiencies, we introduce a Flexible Federated Backdoor Defense Framework (Fedward) to ensure the elimination of adversarial backdoors. We decompose FBA into various attacks, and design amplified magnitude sparsification (AmGrad) and adaptive OPTICS clustering (AutoOPTICS) to address each attack. Meanwhile, Fedward uses the adaptive clipping method by regarding the number of samples in the benign group as constraints on the boundary. This ensures that Fedward can maintain the performance for the Non-IID scenario. We conduct experimental evaluations over three benchmark datasets and thoroughly compare them to state-of-the-art studies. The results demonstrate the promising defense performance from Fedward, moderately improved by 33% $\sim$ 75 in clustering defense methods, and 96.98%, 90.74%, and 89.8% for Non-IID to the utmost extent for the average FBA success rate over MNIST, FMNIST, and CIFAR10, respectively.