Saved in:
Bibliographic Details
Main Authors: Ergisi, Sena, Maßny, Luis, Bitar, Rawad
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.09534
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914032604151808
author Ergisi, Sena
Maßny, Luis
Bitar, Rawad
author_facet Ergisi, Sena
Maßny, Luis
Bitar, Rawad
contents Federated Learning (FL) emerged as a widely studied paradigm for distributed learning. Despite its many advantages, FL remains vulnerable to adversarial attacks, especially under data heterogeneity. We propose a new Byzantine-robust FL algorithm called ProDiGy. The key novelty lies in evaluating the client gradients using a joint dual scoring system based on the gradients' proximity and dissimilarity. We demonstrate through extensive numerical experiments that ProDiGy outperforms existing defenses in various scenarios. In particular, when the clients' data do not follow an IID distribution, while other defense mechanisms fail, ProDiGy maintains strong defense capabilities and model accuracy. These findings highlight the effectiveness of a dual perspective approach that promotes natural similarity among honest clients while detecting suspicious uniformity as a potential indicator of an attack.
format Preprint
id arxiv_https___arxiv_org_abs_2509_09534
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ProDiGy: Proximity- and Dissimilarity-Based Byzantine-Robust Federated Learning
Ergisi, Sena
Maßny, Luis
Bitar, Rawad
Machine Learning
Distributed, Parallel, and Cluster Computing
Federated Learning (FL) emerged as a widely studied paradigm for distributed learning. Despite its many advantages, FL remains vulnerable to adversarial attacks, especially under data heterogeneity. We propose a new Byzantine-robust FL algorithm called ProDiGy. The key novelty lies in evaluating the client gradients using a joint dual scoring system based on the gradients' proximity and dissimilarity. We demonstrate through extensive numerical experiments that ProDiGy outperforms existing defenses in various scenarios. In particular, when the clients' data do not follow an IID distribution, while other defense mechanisms fail, ProDiGy maintains strong defense capabilities and model accuracy. These findings highlight the effectiveness of a dual perspective approach that promotes natural similarity among honest clients while detecting suspicious uniformity as a potential indicator of an attack.
title ProDiGy: Proximity- and Dissimilarity-Based Byzantine-Robust Federated Learning
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2509.09534