Saved in:
Bibliographic Details
Main Authors: Yang, He, Lv, Dongyi, Xi, Wei, Ma, Song, Gu, Hanlin, Zhao, Jizhong
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.15115
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915940627644416
author Yang, He
Lv, Dongyi
Xi, Wei
Ma, Song
Gu, Hanlin
Zhao, Jizhong
author_facet Yang, He
Lv, Dongyi
Xi, Wei
Ma, Song
Gu, Hanlin
Zhao, Jizhong
contents Most existing Byzantine-robust federated learning (FL) methods suffer from slow and unstable convergence. Moreover, when handling a substantial proportion of colluded malicious clients, achieving robustness typically entails compromising model utility. To address these issues, this work introduces FedIDM, which employs distribution matching to construct trustworthy condensed data for identifying and filtering abnormal clients. FedIDM consists of two main components: (1) attack-tolerant condensed data generation, and (2) robust aggregation with negative contribution-based rejection. These components exclude local updates that (1) deviate from the update direction derived from condensed data, or (2) cause a significant loss on the condensed dataset. Comprehensive evaluations on three benchmark datasets demonstrate that FedIDM achieves fast and stable convergence while maintaining acceptable model utility, under multiple state-of-the-art Byzantine attacks involving a large number of malicious clients.
format Preprint
id arxiv_https___arxiv_org_abs_2604_15115
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FedIDM: Achieving Fast and Stable Convergence in Byzantine Federated Learning through Iterative Distribution Matching
Yang, He
Lv, Dongyi
Xi, Wei
Ma, Song
Gu, Hanlin
Zhao, Jizhong
Machine Learning
Cryptography and Security
Most existing Byzantine-robust federated learning (FL) methods suffer from slow and unstable convergence. Moreover, when handling a substantial proportion of colluded malicious clients, achieving robustness typically entails compromising model utility. To address these issues, this work introduces FedIDM, which employs distribution matching to construct trustworthy condensed data for identifying and filtering abnormal clients. FedIDM consists of two main components: (1) attack-tolerant condensed data generation, and (2) robust aggregation with negative contribution-based rejection. These components exclude local updates that (1) deviate from the update direction derived from condensed data, or (2) cause a significant loss on the condensed dataset. Comprehensive evaluations on three benchmark datasets demonstrate that FedIDM achieves fast and stable convergence while maintaining acceptable model utility, under multiple state-of-the-art Byzantine attacks involving a large number of malicious clients.
title FedIDM: Achieving Fast and Stable Convergence in Byzantine Federated Learning through Iterative Distribution Matching
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2604.15115