Work

Over-the-Air Federated Learning with Model Heterogeneity: A Comparative Study

Year: 2024

Type: conference-paper

Abstract: Federated learning (FL) is a promising paradigm that enables collaboration among edge devices to train a neural network while preserving data privacy. This paper considers a previously unexamined scenario of over-the-air FL systems with model heterogeneity, where clients with varying computing capacities adopt local models comprising subsets of global parameters, and over-the-air computation is employed to accelerate parameter aggregation. We investigate three key design considerations, namely, subnet creation, client selection, and client composition, and assess their impact on signal distortion resulting from over-the-air computation and on the accuracy of model learning. Our results provide insights into the effects of each design aspect on the FL system’s performance and convergence.

Source: IEEE Vehicular Technology Conference

DOI: 10.1109/vtc2024-fall63153.2024.10757878

Authors: Yi-Cheng Lai, Ronald Y. Chang, Wei-Yu Chiu

Institutions: Research Center for Information Technology Innovation, Academia Sinica, Academia Sinica, Deakin 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: green

Funder: National Science and Technology Council