Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Efficient Client Selection in Federated Learning

Conference ·
 [1];  [1];  [1]
  1. University of Texas at El Paso,Department of Computer Science,El Paso,USA

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.

Research Organization:
National Energy Technology Laboratory
Sponsoring Organization:
US Department of Energy
DOE Contract Number:
FE0032089
OSTI ID:
2583901
Country of Publication:
United States
Language:
English

References (5)

AdaFL: Adaptive Client Selection and Dynamic Contribution Evaluation for Efficient Federated Learning conference April 2024
Network Anomaly Detection Using Federated Learning conference November 2022
UNSW-NB15: a comprehensive data set for network intrusion detection systems (UNSW-NB15 network data set) conference November 2015
CriticalFL: A Critical Learning Periods Augmented Client Selection Framework for Efficient Federated Learning conference August 2023
A comprehensive guide to CAN IDS data and introduction of the ROAD dataset journal January 2024

Similar Records

Adaptive Client Selection in Federated Learning: A Network Anomaly Detection Use Case
Conference · Sun Feb 16 23:00:00 EST 2025 · OSTI ID:2583897

Reducing Communication Overhead in Federated Learning for Network Anomaly Detection with Adaptive Client Selection
Conference · Mon May 19 00:00:00 EDT 2025 · OSTI ID:2583896

Network Slicing for Federated Learning in Operational Technology Environment
Conference · Mon May 26 00:00:00 EDT 2025 · OSTI ID:3003235

Related Subjects