Article Dans Une Revue ACM Computing Surveys Année : 2025

Bias in Federated Learning: A Comprehensive Survey

Résumé

Federated Learning (FL) enables collaborative model training over multiple clients’ data, without sharing these data for better privacy. Addressing bias in FL remains a challenge. In this paper, we irst present a taxonomy of FL bias, presenting the causes and the diferent types of FL bias, namely demographic bias, performance-related bias, and contribution-related bias. We then categorize FL bias mitigation, in terms of used methods and provided guarantees, before providing a comprehensive and comparative analysis of existing works. Finally, we highlight key challenges and open research directions, including the impact of FL bias mitigation on model utility, privacy, and robustness.

Fichier principal
Vignette du fichier
Bias_in_Federated_Learning_A_Comprehensive_Survey.pdf (845.97 Ko) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte
Licence

Dates et versions

hal-05093158 , version 1 (04-05-2026)

Licence

Identifiants

Citer

Nawel Benarba, Sara Bouchenak. Bias in Federated Learning: A Comprehensive Survey. ACM Computing Surveys, 2025, 57 (11), pp.291. ⟨10.1145/3735125⟩. ⟨hal-05093158⟩
159 Consultations
0 Téléchargements

Altmetric

Partager

  • More