RADAR: Model Quality Assessment for Reputation-aware Collaborative Federated Learning - Archive ouverte HAL
Communication Dans Un Congrès Année : 2024

RADAR: Model Quality Assessment for Reputation-aware Collaborative Federated Learning

Résumé

Cross-silo federated learning (CS-FL) is a distributed learning setting which allows an identified set of organizations to collaboratively train a single global model. Since CS-FL use cases are often heterogeneous, it may be more appropriate to dynamically provide different models to more homogeneous subfederations. In addition, such systems can be undermined by contributions of poor quality, making negligent or even malicious participants critical to consider. However, distinguishing such participants in a heterogeneous context is especially difficult. We present RADAR, a novel architecture for CS-FL able to assess the quality of the participants' contributions, regardless of data similarity. RADAR leverages client-side evaluation to directly collect feedbacks from the participants. The same evaluations allow grouping participants according to their perceived similarity and weighting the model aggregation based on their reputation. To evaluate our approach on concrete experiments, we implement a collaborative intrusion detection system (CIDS) scenario and test our architecture in various data-quality settings using labelflipping. Our results confirm that combining clustering and a reputation system succeeds in detecting a wide range of Byzantine behaviors, including colluding attackers, which highlights RADAR's versatility.
Fichier principal
Vignette du fichier
Trustfids_mitigation-3.pdf (476.61 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04681789 , version 1 (30-08-2024)

Identifiants

  • HAL Id : hal-04681789 , version 1

Citer

Léo Lavaur, Pierre-Marie Lechevalier, Yann Busnel, Romaric Ludinard, Géraldine Texier, et al.. RADAR: Model Quality Assessment for Reputation-aware Collaborative Federated Learning. 43rd International Symposium on Reliable Distributed Systems - SRDS 2024, Sep 2024, Charlotte (NC), United States. ⟨hal-04681789⟩
49 Consultations
37 Téléchargements

Partager

More