Publications

BayBFed: Bayesian Backdoor Defense for Federated Learning

AuthorKumari, Kavita; Rieger, Phillip; Fereidooni, Hossein; Jadliwala, Murtuza; Sadeghi, Ahmad-Reza
Date2023
TypeConference Proceedings
AbstractFederated learning (FL) is an emerging technology that allows participants to jointly train a machine learning model without sharing their private data with others. However, FL is vulnerable to poisoning attacks such as backdoor attacks. Consequently, a variety of defenses have recently been proposed, which have primarily utilized intermediary states of the global model (i.e., logits) or distance of the local models (i.e., L2−norm) with respect to the global model to detect malicious backdoors in FL. However, as these approaches directly operate on client updates (or weights), their effectiveness depends on factors such as clients’ data distribution or the adversary’s attack strategies. In this paper, we introduce a novel and more generic backdoor defense framework, called BayBFed, which proposes to utilize probability distributions over client updates to detect malicious updates in FL: BayBFed computes a probabilistic measure over the clients’ updates to keep track of any adjustments made in the updates, and uses a novel detection algorithm that can leverage this probabilistic measure to efficiently detect and filter out malicious updates. Thus, it overcomes the shortcomings of previous approaches that arise due to the direct usage of client updates; nevertheless, our probabilistic measure will include all aspects of the local client training strategies. BayBFed utilizes two Bayesian Non-Parametric (BNP) extensions: (i) a Hierarchical Beta-Bernoulli process to draw a probabilistic measure given the clients’ updates, and (ii) an adaptation of the Chinese Restaurant Process (CRP), referred by us as CRP-Jensen, which leverages this probabilistic measure to detect and filter out malicious updates. We extensively evaluate our defense approach on five benchmark datasets: CIFAR10, Reddit, IoT intrusion detection, MNIST, and FMNIST, and show that it can effectively detect and eliminate malicious updates in FL without deteriorating the benign performance of the global model.
Conference44th IEEE Symposium on Security and Privacy
Isbn978-1-6654-9336-9
InProceedings: 44th IEEE Symposium on Security and Privacy - SP 2023, p.747-764
PublisherIEEE
Urlhttps://tubiblio.ulb.tu-darmstadt.de/id/eprint/136834