FLASH: Heterogeneity-Aware Federated Learning at Scale

Advanced search
Browse by:

Department | Year

UCL Theses | Latest

Deposit your research

FLASH: Heterogeneity-Aware Federated Learning at Scale

Yang, C; Xu, M; Wang, Q; Chen, Z; Huang, K; Ma, Y; Bian, K; ... Liu, X; + view all (2022) FLASH: Heterogeneity-Aware Federated Learning at Scale. IEEE Transactions on Mobile Computing 10.1109/TMC.2022.3214234. (In press). Green open access

Preview

Text
tmc22.pdf - Accepted Version
Download (3MB) | Preview

Abstract

Federated learning (FL) becomes a promising machine learning paradigm. The impact of heterogeneous hardware specifications and dynamic states on the FL process has not yet been studied systematically. This paper presents the first large-scale study of this impact based on real-world data collected from 136k smartphones. We conducted extensive experiments on our proposed heterogeneity-aware FL platform namely FLASH , to systematically explore the performance of state-of-the-art FL algorithms and key FL configurations in heterogeneity-aware and -unaware settings, finding the following. (1) Heterogeneity causes accuracy to drop by up to 9.2% and convergence time to increase by 2.32×. (2) Heterogeneity negatively impacts popular aggregation algorithms, e.g., the accuracy variance reduction brought by q-FedAvg drops by 17.5%. (3) Heterogeneity does not worsen the accuracy loss caused by gradient-compression algorithms significantly, but it compromises the convergence time by up to 2.5×. (4) Heterogeneity hinders client-selection algorithms from selecting wanted clients, thus reducing effectiveness. e.g., the accuracy increase brought by the state-of-the-art client-selection algorithm drops by 73.9%. (5) Heterogeneity causes the optimal FL hyper-parameters to drift significantly. More specifically, the heterogeneity-unaware setting favors looser deadline and higher reporting fraction to achieve better training performance. (6) Heterogeneity results in non-trivial failed clients (more than 10%) and leads to participation bias (the top 30% of clients contribute 86% of computations). Our FLASH platform and data have been publicly open sourced.

Type:	Article
Title:	FLASH: Heterogeneity-Aware Federated Learning at Scale
Open access status:	An open access version is available from UCL Discovery
DOI:	10.1109/TMC.2022.3214234
Publisher version:	https://doi.org/10.1109/TMC.2022.3214234
Language:	English
Additional information:	This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords:	Federated learning, heterogeneity, impact analysis
UCL classification:	UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
URI:	https://discovery.ucl.ac.uk/id/eprint/10158818