Random matrix improved subspace clustering

Handle URI:
http://hdl.handle.net/10754/623206
Title:
Random matrix improved subspace clustering
Authors:
Couillet, Romain; Kammoun, Abla ( 0000-0002-0195-3159 )
Abstract:
This article introduces a spectral method for statistical subspace clustering. The method is built upon standard kernel spectral clustering techniques, however carefully tuned by theoretical understanding arising from random matrix findings. We show in particular that our method provides high clustering performance while standard kernel choices provably fail. An application to user grouping based on vector channel observations in the context of massive MIMO wireless communication networks is provided.
KAUST Department:
King Abdullah University of Science and Technology, Saudi Arabia
Citation:
Couillet R, Kammoun A (2016) Random matrix improved subspace clustering. 2016 50th Asilomar Conference on Signals, Systems and Computers. Available: http://dx.doi.org/10.1109/ACSSC.2016.7869000.
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2016 50th Asilomar Conference on Signals, Systems and Computers
Conference/Event name:
50th Asilomar Conference on Signals, Systems and Computers, ACSSC 2016
Issue Date:
6-Mar-2017
DOI:
10.1109/ACSSC.2016.7869000
Type:
Conference Paper
Sponsors:
Couillet’s work is supported by the ANR Project RMT4GRAPH (ANR-14-CE28-0006).
Additional Links:
http://ieeexplore.ieee.org/document/7869000/
Appears in Collections:
Conference Papers

Full metadata record

DC FieldValue Language
dc.contributor.authorCouillet, Romainen
dc.contributor.authorKammoun, Ablaen
dc.date.accessioned2017-04-13T11:51:01Z-
dc.date.available2017-04-13T11:51:01Z-
dc.date.issued2017-03-06en
dc.identifier.citationCouillet R, Kammoun A (2016) Random matrix improved subspace clustering. 2016 50th Asilomar Conference on Signals, Systems and Computers. Available: http://dx.doi.org/10.1109/ACSSC.2016.7869000.en
dc.identifier.doi10.1109/ACSSC.2016.7869000en
dc.identifier.urihttp://hdl.handle.net/10754/623206-
dc.description.abstractThis article introduces a spectral method for statistical subspace clustering. The method is built upon standard kernel spectral clustering techniques, however carefully tuned by theoretical understanding arising from random matrix findings. We show in particular that our method provides high clustering performance while standard kernel choices provably fail. An application to user grouping based on vector channel observations in the context of massive MIMO wireless communication networks is provided.en
dc.description.sponsorshipCouillet’s work is supported by the ANR Project RMT4GRAPH (ANR-14-CE28-0006).en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.relation.urlhttp://ieeexplore.ieee.org/document/7869000/en
dc.subjectClustering algorithmsen
dc.subjectCorrelationen
dc.subjectEigenvalues and eigenfunctionsen
dc.subjectKernelen
dc.subjectLaplace equationsen
dc.subjectMIMOen
dc.subjectStandardsen
dc.titleRandom matrix improved subspace clusteringen
dc.typeConference Paperen
dc.contributor.departmentKing Abdullah University of Science and Technology, Saudi Arabiaen
dc.identifier.journal2016 50th Asilomar Conference on Signals, Systems and Computersen
dc.conference.date2016-11-06 to 2016-11-09en
dc.conference.name50th Asilomar Conference on Signals, Systems and Computers, ACSSC 2016en
dc.conference.locationPacific Grove, CA, USAen
dc.contributor.institutionLANEAS Group, CentraleSup´elec, University of Paris-Saclay, Franceen
kaust.authorKammoun, Ablaen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.