Efficient Client Selection in Federated Learning

William Marfo; Deepak K. Tosh; Shirley V. Moore

Efficient Client Selection in Federated Learning

Machine Learning 2025-02-04 v1 Artificial Intelligence Distributed, Parallel, and Cluster Computing

Authors: William Marfo , Deepak K. Tosh , Shirley V. Moore

Abstract

Federated Learning (FL) enables decentralized machine learning while preserving data privacy. This paper proposes a novel client selection framework that integrates differential privacy and fault tolerance. The adaptive client selection adjusts the number of clients based on performance and system constraints, with noise added to protect privacy. Evaluated on the UNSW-NB15 and ROAD datasets for network anomaly detection, the method improves accuracy by 7% and reduces training time by 25% compared to baselines. Fault tolerance enhances robustness with minimal performance trade-offs.

Keywords

federated learning federated graph learning distributed learning

Cite

@article{arxiv.2502.00036,
  title  = {Efficient Client Selection in Federated Learning},
  author = {William Marfo and Deepak K. Tosh and Shirley V. Moore},
  journal= {arXiv preprint arXiv:2502.00036},
  year   = {2025}
}

Related papers

View all related →