Backdoor Attacks in Peer-to-Peer Federated Learning
ACM Transactions on Privacy and Security · 2025-02-28 · 28 (1) · 1-28
DOI: 10.1145/3691633 ↗At a glance
Journal article in ACM Transactions on Privacy and Security, published online on 2024-12-11. Written by Georgios Syros and 4 co-authors. Cited 7 times so far (Crossref); 99 references. The full text is freely available (open access, CC BY-SA 4.0). Funded by: Department of Defense Multidisciplinary Research Program of the University Research Initiative.
Abstract (original language)
Most machine learning applications rely on centralized learning processes, opening up the risk of exposure of their training datasets. While federated learning (FL) mitigates to some extent these privacy risks, it relies on a trusted aggregation server for training a shared global model. Recently, new distributed learning architectures based on Peer-to-Peer Federated Learning (P2PFL) offer advantages in terms of both privacy and reliability. Still, their resilience to poisoning attacks during training has not been investigated. In this article, we propose new backdoor attacks for P2PFL that leverage structural graph properties to select the malicious nodes, and achieve high attack success, while remaining stealthy. We evaluate our attacks under various realistic conditions, including multiple graph topologies, limited adversarial visibility of the network, and clients with non-IID data. Finally, we show the limitations of existing defenses adapted from FL and design a new defense that successfully mitigates the backdoor attacks, without an impact on model accuracy.
Funding
- Department of Defense Multidisciplinary Research Program of the University Research Initiative
Cite
APA
Syros, G., Yar, G., Boboila, S., Nita-Rotaru, C., & Oprea, A. (2025). Backdoor Attacks in Peer-to-Peer Federated Learning. ACM Transactions on Privacy and Security, 28(1), 1-28. https://doi.org/10.1145/3691633
Vancouver
Syros G, Yar G, Boboila S, Nita-Rotaru C, Oprea A. Backdoor Attacks in Peer-to-Peer Federated Learning. ACM Transactions on Privacy and Security. 2025;28(1):1-28. doi:10.1145/3691633
BibTeX
@article{Syros2025,
author = {Syros, Georgios and Yar, Gokberk and Boboila, Simona and Nita-Rotaru, Cristina and Oprea, Alina},
title = {Backdoor Attacks in Peer-to-Peer Federated Learning},
journal = {ACM Transactions on Privacy and Security},
year = {2025},
volume = {28},
number = {1},
pages = {1--28},
doi = {10.1145/3691633}
}
EUVIMED