The Role of Communication Time in the Convergence of Federated Edge Learning

Zhou, Y; Fu, Y; Luo, Z; Hu, M; Wu, D; Sheng, QZ; Yu, S

The Role of Communication Time in the Convergence of Federated Edge Learning

Zhou, Y Fu, Y Luo, Z Hu, M Wu, D Sheng, QZ Yu, S

Permalink

Publisher:: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Publication Type:: Journal Article
Citation:: IEEE Transactions on Vehicular Technology, 2022, 71, (3), pp. 3241-3254
Issue Date:: 2022-03-01

Closed Access

	Filename	Description	Size
	The_Role_of_Communication_Time_in_the_Convergence_of_Federated_Edge_Learning.pdf		1.38 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Zhou, Y
dc.contributor.author	Fu, Y
dc.contributor.author	Luo, Z
dc.contributor.author	Hu, M
dc.contributor.author	Wu, D
dc.contributor.author	Sheng, QZ
dc.contributor.author	Yu, S https://orcid.org/0000-0003-4485-6743
dc.date.accessioned	2023-04-06T03:01:55Z
dc.date.available	2023-04-06T03:01:55Z
dc.date.issued	2022-03-01
dc.identifier.citation	IEEE Transactions on Vehicular Technology, 2022, 71, (3), pp. 3241-3254
dc.identifier.issn	0018-9545
dc.identifier.issn	1939-9359
dc.identifier.uri	http://hdl.handle.net/10453/169280
dc.description.abstract	Federated Edge Learning (FEL) enables a massive number of edge devices (e.g. smart phones) to train machine learning models collaboratively. Due to the inherent unreliability of participating edge devices and the unpredictable network conditions, the final model accuracy is largely determined by the communication time between edge devices and the parameter server (PS). However, there exists very limited work to quantify the influence of the communication time in FEL and show how to minimize its negative impacts from a theoretic perspective. In this paper, we are among the first to develop a formal model of the communication time in FEL and its influence on the final model accuracy. In our work, the set of edge devices involved in each global iteration is defined as the ECP (Engaged Client Pool). We model the communication time cost as a function with respect to the response time distribution of individual devices and the ECP size. By incorporating communication time cost into the convergence rate analysis, we propose the ECPA and H-ECPA algorithms to automatically adjust the size of the ECP so as to maximize the model accuracy in both homogeneous and heterogeneous networks. We also analyze how the tail shape of response time affects the convergence rate, and prove that the heavy tail shape can significantly lower the model accuracy. Finally, we conduct extensive experiments with real datasets, and the results confirm the correctness of our analysis and demonstrate the superiority of our proposed algorithms.
dc.language	English
dc.publisher	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
dc.relation	http://purl.org/au-research/grants/arc/DP200101374
dc.relation	http://purl.org/au-research/grants/arc/LP190100676
dc.relation.ispartof	IEEE Transactions on Vehicular Technology
dc.relation.isbasedon	10.1109/TVT.2022.3144099
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	08 Information and Computing Sciences, 09 Engineering, 10 Technology
dc.subject.classification	Automobile Design & Engineering
dc.title	The Role of Communication Time in the Convergence of Federated Edge Learning
dc.type	Journal Article
utslib.citation.volume	71
utslib.for	08 Information and Computing Sciences
utslib.for	09 Engineering
utslib.for	10 Technology
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	closed_access	*
dc.date.updated	2023-04-06T03:01:54Z
pubs.issue	3
pubs.publication-status	Published
pubs.volume	71
utslib.citation.issue	3

Abstract:

Federated Edge Learning (FEL) enables a massive number of edge devices (e.g. smart phones) to train machine learning models collaboratively. Due to the inherent unreliability of participating edge devices and the unpredictable network conditions, the final model accuracy is largely determined by the communication time between edge devices and the parameter server (PS). However, there exists very limited work to quantify the influence of the communication time in FEL and show how to minimize its negative impacts from a theoretic perspective. In this paper, we are among the first to develop a formal model of the communication time in FEL and its influence on the final model accuracy. In our work, the set of edge devices involved in each global iteration is defined as the ECP (Engaged Client Pool). We model the communication time cost as a function with respect to the response time distribution of individual devices and the ECP size. By incorporating communication time cost into the convergence rate analysis, we propose the ECPA and H-ECPA algorithms to automatically adjust the size of the ECP so as to maximize the model accuracy in both homogeneous and heterogeneous networks. We also analyze how the tail shape of response time affects the convergence rate, and prove that the heavy tail shape can significantly lower the model accuracy. Finally, we conduct extensive experiments with real datasets, and the results confirm the correctness of our analysis and demonstrate the superiority of our proposed algorithms.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/169280