mlsanityData to 7 Oct 2026

FedIDM: Achieving Fast and Stable Convergence in Byzantine Federated Learning through Iterative Distribution Matching

arXiv 2604.15115 ↗First posted 16 Apr 2026cs.LG cs.CRcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 37 references that are in the map. First posted on arXiv 16 Apr 2026.

Builds on

25 of 37 references.

References · 25

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  3. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  4. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  5. 2019Advances and Open Problems in Federated Learningcited by 3,021
  6. 2021Model-Contrastive Federated Learningcited by 762
  7. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 698
  8. 2020Federated Learning with Matched Averagingcited by 564
  9. 2019Local Model Poisoning Attacks to Byzantine-Robust Federated Learningcited by 479
  10. 2021Dataset Condensation with Distribution Matchingcited by 362
  11. 2020FLTrust: Byzantine-robust Federated Learning via Trust Bootstrappingcited by 330
  12. 2018The Hidden Vulnerability of Distributed Learning in Byzantiumcited by 324
13 more
  1. 2019A Little Is Enough: Circumventing Defenses For Distributed Learningcited by 285
  2. 2020Privacy and Robustness in Federated Learning: Attacks and Defensescited by 163
  3. 2023Improved Distribution Matching for Dataset Condensationcited by 124
  4. 2022FLDetector: Defending Federated Learning Against Model Poisoning Attacks via Detecting Malicious Clientscited by 119
  5. 2021Back to the Drawing Board: A Critical Evaluation of Poisoning Attacks on Production Federated Learningcited by 116
  6. 2023Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmarkcited by 97
  7. 2022FedDM: Iterative Distribution Matching for Communication-Efficient Federated Learningcited by 65
  8. 2023An Experimental Study of Byzantine-Robust Aggregation Schemes in Federated Learningcited by 32
  9. 2023A Huber Loss Minimization Approach to Byzantine Robust Federated Learningcited by 16
  10. 2022FL-Defender: Combating Targeted Attacks in Federated Learningcited by 12
  11. 2023FedDefender: Client-Side Attack-Tolerant Federated Learningcited by 12
  12. 2023A Four-Pronged Defense Against Byzantine Attacks in Federated Learningcited by 9
  13. 2024BadSampler: Harnessing the Power of Catastrophic Forgetting to Poison Byzantine-robust Federated Learningcited by 3

Built on by

Not cited yet.