Work

Poster: Delay and Energy-Efficient Client Selection for Federated Learning in Vehicular Networks

Year: 2024

Type: conference-paper

Abstract: This paper proposes a delay and energy-aware clustering-based client selection scheme for federated learning in vehicular networks. We propose an algorithm that selects the appropriate number of vehicles for local training, minimizing delay and energy consumption while ensuring model performance. The simulation results demonstrate that the proposed algorithm achieves lower delay and energy consumption compared to benchmark methods, for both IID (independent and identically distributed) and non-IID datasets.

Source: ACM Conference on Embedded Networked Sensor Systems (SenSys)

DOI: 10.1145/3666025.3699399

Authors: Junho Jeong, Chang Kyung Kim, SuKyoung Lee

Institution: Yonsei University

Language: English

Topic: Privacy-Preserving Technologies in Data

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: bronze

Funders: National Research Foundation, National Research Foundation of Korea, Ministry of Science and ICT, South Korea