Ajou University repository

Addressing Client Heterogeneity in Synchronous Federated Learning: The CHAFL Approach
Citations

SCOPUS

0

Citation Export

Publication Year
2023-01-01
Journal
Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS
Publisher
IEEE Computer Society
Citation
Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS, pp.2804-2805
Keyword
adaptive local epochclient selectionfederated learningsystems and statistical heterogeneity
Mesh Keyword
Adaptive local epochClient selectionFederated learningGlobal modelsLearning approachLocal lossSecurity vulnerabilitiesStatistical heterogeneitiesSystem and statistical heterogeneitySystem aspects
All Science Classification Codes (ASJC)
Hardware and Architecture
Abstract
Federated learning (FL) is proposed to address the security vulnerabilities of conventional distributed deep learning. Since the capabilities of participating FL clients are highly variable in terms of both statistical and system aspects, FL training will face diminished convergence accuracy and speed. Hence, we propose CHAFL (client heterogeneity aware federated learning) to address client heterogeneity (i.e., statistical and system heterogeneity) in synchronous FL. CHAFL selects clients based on global loss with contribution, defined as local loss, enabling higher round-to-accuracy for the global model than previous studies. Additionally, to handle system heterogeneity, it proposes a lightweight algorithm that eliminates the profiling process previously employed to calculate adaptive local epochs in existing studies, thereby improving time-to-accuracy. In order to verify the effectiveness of CHAFL, we exploit three benchmark datasets on non-IID and system heterogeneous setting for empirical evaluation. Compared to the baseline, CHAFL achieves an accuracy improvement of 3.1 to 5.7%, along with 1.04 to 1.73x higher round-to-accuracy and 1.03 to 1.98x higher time-to-accuracy.
ISSN
1521-9097
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/36956
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85190267358&origin=inward
DOI
https://doi.org/10.1109/icpads60453.2023.00387
Type
Conference
Funding
This work was jointly supported by the Basic Science Research Program of National Research Foundation of Korea (2021R1F1A1062779), the MSIT(Ministry of Science and ICT), Korea, under the ITRC(Information Technology Research Center) support program(IITP-2023-2018-0- 01431) supervised by the IITP(Institute for Information & Communications Technology Planning & Evaluation), and the Korea Institute of Science and Technology Information (KISTI) (KSC2022-CRE-0406).
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Oh, Sangyoon Image
Oh, Sangyoon오상윤
Department of Software and Computer Engineering
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.