Hiding in the crowd: federated data augmentation for on-device learning

Jeong, Eunjeong, Oh, Seungeun, Park, Jihong, Kim, Hyesung, Bennis, Mehdi and Kim, Seong-Lyun 2020, Hiding in the crowd: federated data augmentation for on-device learning, IEEE intelligent systems, pp. 1-8, doi: 10.1109/MIS.2020.3028613.

Attached Files
Name Description MIMEType Size Downloads

Title Hiding in the crowd: federated data augmentation for on-device learning
Author(s) Jeong, Eunjeong
Oh, Seungeun
Park, JihongORCID iD for Park, Jihong orcid.org/0000-0001-7623-6552
Kim, Hyesung
Bennis, Mehdi
Kim, Seong-Lyun
Journal name IEEE intelligent systems
Start page 1
End page 8
Total pages 8
Publisher Institute of Electrical and Electronics Engineers
Place of publication Piscataway, N.J.
Publication date 2020-10-12
ISSN 1541-1672
1941-1294
Keyword(s) Machine learning
Distributed networks
Distributed artificial intelligence
Wireless communication
Summary To cope with the lack of on-device machine learning samples, this article presents a distributed data augmentation algorithm, coined federated data augmentation (FAug). In FAug, devices share a tiny fraction of their local data, i.e., seed samples, and collectively train a synthetic sample generator that can augment the local datasets of devices. To further improve FAug, we introduce a multi-hop based seed sample collection method and an oversampling technique that mixes up collected seed samples. Both approaches enjoy the benefit from the crowd of devices, by hiding data privacy from preceding hops and feeding diverse seed samples. In the image classification tasks, simulations demonstrate that the proposed FAug frameworks yield stronger privacy guarantees, lower communication latency, and higher on-device ML accuracy.
Language eng
DOI 10.1109/MIS.2020.3028613
Indigenous content off
Field of Research 0801 Artificial Intelligence and Image Processing
0806 Information Systems
0906 Electrical and Electronic Engineering
HERDC Research category C1 Refereed article in a scholarly journal
Persistent URL http://hdl.handle.net/10536/DRO/DU:30144500

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 1 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 13 Abstract Views, 3 File Downloads  -  Detailed Statistics
Created: Mon, 26 Oct 2020, 17:08:20 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.