Privacy Inference-Empowered Stealthy Backdoor Attack on Federated Learning under Non-IID Scenarios
This addresses security vulnerabilities in federated learning systems for applications like distributed machine learning, but it is incremental as it builds on existing backdoor attack methods.
The paper tackles the problem of backdoor attacks becoming less effective in federated learning under non-IID data scenarios by proposing a privacy inference-empowered stealthy backdoor attack (PI-SBA) scheme, which uses GANs to reconstruct data and improve attack effectiveness, achieving demonstrated stealthiness against state-of-the-art defenses on datasets like MNIST, CIFAR10, and Youtube Aligned Face.
Federated learning (FL) naturally faces the problem of data heterogeneity in real-world scenarios, but this is often overlooked by studies on FL security and privacy. On the one hand, the effectiveness of backdoor attacks on FL may drop significantly under non-IID scenarios. On the other hand, malicious clients may steal private data through privacy inference attacks. Therefore, it is necessary to have a comprehensive perspective of data heterogeneity, backdoor, and privacy inference. In this paper, we propose a novel privacy inference-empowered stealthy backdoor attack (PI-SBA) scheme for FL under non-IID scenarios. Firstly, a diverse data reconstruction mechanism based on generative adversarial networks (GANs) is proposed to produce a supplementary dataset, which can improve the attacker's local data distribution and support more sophisticated strategies for backdoor attacks. Based on this, we design a source-specified backdoor learning (SSBL) strategy as a demonstration, allowing the adversary to arbitrarily specify which classes are susceptible to the backdoor trigger. Since the PI-SBA has an independent poisoned data synthesis process, it can be integrated into existing backdoor attacks to improve their effectiveness and stealthiness in non-IID scenarios. Extensive experiments based on MNIST, CIFAR10 and Youtube Aligned Face datasets demonstrate that the proposed PI-SBA scheme is effective in non-IID FL and stealthy against state-of-the-art defense methods.