DBA-DFL: Towards Distributed Backdoor Attacks with Network Detection in Decentralized Federated Learning

TL;DR

This paper explores distributed backdoor attacks in decentralized federated learning, proposing a network detection and clustering method to enhance attack success rates regardless of attacker placement.

Contribution

It introduces a novel approach to organize attackers based on network distance and dynamically embed patterns, improving attack effectiveness in decentralized FL.

Findings

Outperforms centralized attacks in decentralized settings

Effective detection and clustering of attackers based on network distance

Achieves higher attack success rates across benchmark datasets

Abstract

Distributed backdoor attacks (DBA) have shown a higher attack success rate than centralized attacks in centralized federated learning (FL). However, it has not been investigated in the decentralized FL. In this paper, we experimentally demonstrate that, while directly applying DBA to decentralized FL, the attack success rate depends on the distribution of attackers in the network architecture. Considering that the attackers can not decide their location, this paper aims to achieve a high attack success rate regardless of the attackers' location distribution. Specifically, we first design a method to detect the network by predicting the distance between any two attackers on the network. Then, based on the distance, we organize the attackers in different clusters. Lastly, we propose an algorithm to \textit{dynamically} embed local patterns decomposed from a global pattern into the different attackers in each cluster. We conduct a thorough empirical investigation and find that our method can, in benchmark datasets, outperform both centralized attacks and naive DBA in different decentralized frameworks.