Divergence-based adaptive aggregation for Byzantine robust federated learning
Author Identifier (ORCID)
Abstract
Inherent client drifts caused by data heterogeneity, as well as vulnerability to Byzantine attacks within the system, hinder effective model training and convergence in federated learning (FL). This paper presents two new frameworks, named DiveRgence-based Adaptive aGgregation (DRAG) and Byzantine-Resilient DRAG (BR-DRAG), to mitigate client drifts and resist attacks while expediting training. DRAG designs a reference direction and a metric named divergence of degree to quantify the deviation of local updates. Accordingly, each worker can align its local update via linear calibration without extra communication cost. BR-DRAG refines DRAG under Byzantine attacks by maintaining a vetted root dataset at the server to produce trusted reference directions. The workers' updates can be then calibrated to mitigate divergence caused by malicious attacks. We analytically prove that DRAG and BR-DRAG achieve fast convergence for non-convex models under partial worker participation, data heterogeneity, and Byzantine attacks. Experiments validate the effectiveness of DRAG and its superior performance over state-of-the-art methods in handling client drifts, and highlight the robustness of BR-DRAG in maintaining resilience against data heterogeneity and diverse Byzantine attacks.
Keywords
byzantine attack, client drift, convergence analysis, federated learning
Document Type
Journal Article
Date of Publication
1-1-2026
E-ISSN
15566021
ISSN
15566013
Volume
21
Publication Title
IEEE Transactions on Information Forensics and Security
Publisher
IEEE
School
School of Engineering
RAS ID
101828
Funding Information
Innovation Program of Shanghai Municipal Science and Technology Commission (Grant Number: 25DP1500300), National Natural Science Foundation of China (Grant Number: 62231010).
Copyright
subscription content
First Page
6284
Last Page
6299
Recommended Citation
Xiao, B., Zhu, F., Zhang, J., Ni, W., & Wang, X. (2026). Divergence-based adaptive aggregation for Byzantine robust federated learning. IEEE Transactions on Information Forensics and Security, 21, 6284–6299. https://doi.org/10.1109/TIFS.2026.3707437