Byzantine-robust Decentralized Federated Learning via Dual-domain Clustering and Trust Bootstrapping
Authors:
- 16 June 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 24756-24765
Abstract
Decentralized federated learning (DFL) facilitates collaborative model training across multiple connected clients without a central coordination server, thereby avoiding the single point of failure in traditional centralized federated learning (CFL). However, DFL exhibits increased susceptibility to Byzantine attacks owing to the lack of a responsible central server. Furthermore, a benign client in DFL may be dominated by Byzantine clients (more than half of its neighbors are malicious), posing significant challenges for robust model training. In this work, we propose DFL-Dual, a novel Byzantine-robust DFL method through dual-domain client clustering and trust bootstrapping. Specifically, we first propose to leverage both data-domain and model-domain distance metrics to identify client discrepancies. Then, we design a trust evaluation mechanism centered on benign clients, which enables them to evaluate their neighbors. Building upon the dual-domain distance metric and trust evaluation mechanism, we further develop a two-stage clustering and trust bootstrapping technique to exclude Byzantine clients from local model aggregation. We extensively evaluate the proposed DFL-Dual method through rigorous experimentation, demonstrating its remarkable performance superiority over existing robust CFL and DFL schemes.Keywords
Funding Information
- National Natural Science Foundation of China (62102337,62203309,62122066,U20A20182)
- Natural Science Foundation of Hunan Province (2023JJ40174)
- Fundamental Research Funds for the Central Universities
This publication has 25 references indexed in Scilit:
- Multi-metrics adaptively identifies backdoors in Federated learning2023
- Decentralized Federated Learning: Fundamentals, State of the Art, Frameworks, Trends, and ChallengesIEEE Communications Surveys & Tutorials, 2023
- Make Landscape Flatter in Differentially Private Federated Learning2023
- Federated Learning with Data-Agnostic Distribution Fusion2023
- The Resource Problem of Using Linear Layer Leakage Attack in Federated Learning2023
- CaPriDe Learning: Confidential and Private Decentralized Learning Based on Encryption-Friendly Distillation Loss2023
- Decentralized federated learning through proxy model sharingNature Communications, 2023
- Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakage2022
- Semi-Decentralized Federated Learning With Cooperative D2D Local Model AggregationsIEEE Journal on Selected Areas in Communications, 2021
- A consensus-based decentralized training algorithm for deep neural networks with communication compressionNeurocomputing, 2021