dc.contributor.author | Mills, J | |
dc.contributor.author | Hu, J | |
dc.contributor.author | Min, G | |
dc.date.accessioned | 2019-11-27T10:22:53Z | |
dc.date.issued | 2019-11-28 | |
dc.description.abstract | The rapidly expanding number of IoT devices is
generating huge quantities of data, but public concern over
data privacy means users are apprehensive to send data to a
central server for Machine Learning (ML) purposes. The easilychanged behaviours of edge infrastructure that Software Defined
Networking provides makes it possible to collate IoT data at
edge servers and gateways, where Federated Learning (FL) can
be performed: building a central model without uploading data to
the server. FedAvg is a FL algorithm which has been the subject
of much study, however it suffers from a large number of rounds
to convergence with non-Independent, Identically Distributed
(non-IID) client datasets and high communication costs per
round. We propose adapting FedAvg to use a distributed form
of Adam optimisation, greatly reducing the number of rounds
to convergence, along with novel compression techniques, to
produce Communication-Efficient FedAvg (CE-FedAvg). We perform extensive experiments with the MNIST/CIFAR-10 datasets,
IID/non-IID client data, varying numbers of clients, client participation rates, and compression rates. These show CE-FedAvg
can converge to a target accuracy in up to 6× less rounds than
similarly compressed FedAvg, while uploading up to 3× less data,
and is more robust to aggressive compression. Experiments on
an edge-computing-like testbed using Raspberry Pi clients also
show CE-FedAvg is able to reach a target accuracy in up to 1.7×
less real time than FedAvg. | en_GB |
dc.description.sponsorship | Engineering and Physical Sciences Research Council (EPSRC) | en_GB |
dc.identifier.citation | Publshed online 28 November 2019 | en_GB |
dc.identifier.doi | 10.1109/JIOT.2019.2956615 | |
dc.identifier.uri | http://hdl.handle.net/10871/39842 | |
dc.language.iso | en | en_GB |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | en_GB |
dc.rights | © 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. | |
dc.subject | Federated Learning | en_GB |
dc.subject | Internet of Things | en_GB |
dc.subject | Distributed Computing | en_GB |
dc.subject | Edge Computing | en_GB |
dc.subject | Compression | en_GB |
dc.title | Communication-Efficient Federated Learning for Wireless Edge Intelligence in IoT | en_GB |
dc.type | Article | en_GB |
dc.date.available | 2019-11-27T10:22:53Z | |
dc.identifier.issn | 2327-4662 | |
dc.description | This is the author accepted manuscript. The final version is available from IEEE via the DOI in this record | en_GB |
dc.identifier.journal | IEEE Internet of Things | en_GB |
dc.rights.uri | http://www.rioxx.net/licenses/all-rights-reserved | en_GB |
dcterms.dateAccepted | 2019-11-21 | |
exeter.funder | ::Engineering and Physical Sciences Research Council (EPSRC) | en_GB |
rioxxterms.version | AM | en_GB |
rioxxterms.licenseref.startdate | 2019-11-21 | |
rioxxterms.type | Journal Article/Review | en_GB |
refterms.dateFCD | 2019-11-26T22:26:45Z | |
refterms.versionFCD | AM | |
refterms.dateFOA | 2019-12-02T16:19:32Z | |
refterms.panel | B | en_GB |