A Fast Generalized Low Rank Representation Framework Based on L<inf>2,p</inf> Norm Minimization for Subspace Clustering

Zhang, T; Tang, Z; Shen, X

A Fast Generalized Low Rank Representation Framework Based on L<inf>2,p</inf> Norm Minimization for Subspace Clustering

Zhang, T

Tang, Z Shen, X

Permalink

Publication Type:: Journal Article
Citation:: IEEE Access, 2017, 5 pp. 23299 - 23311
Issue Date:: 2017-10-21

Closed Access

	Filename	Description	Size
	08080223.pdf	Published Version	8.47 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	Zhang, T https://orcid.org/0000-0003-2673-9914	en_US
dc.contributor.author	Tang, Z	en_US
dc.contributor.author	Shen, X https://orcid.org/0000-0001-8494-4532	en_US
dc.date.issued	2017-10-21	en_US
dc.identifier.citation	IEEE Access, 2017, 5 pp. 23299 - 23311	en_US
dc.identifier.uri	http://hdl.handle.net/10453/126817
dc.description.abstract	© 2017 IEEE. Low rank representation (LRR) is powerful for subspace clustering due to its strong ability in exploring low-dimensional subspace structures embedded in data. LRR is usually solved by iterative nuclear norm minimization, which involves singular value decomposition (SVD) at each iteration. However, the multiple SVDs limit the application of LRR due to its high computational cost. In this paper, we propose fast generalized LRR to address the above issue. Specifically, the nuclear norm and L2,1 norm in LRR are generalized to be the Schatten-p norm and L2,q norm, respectively. The new model is more general and robust than LRR. Then, we decompose the data matrix by Qatar riyal decomposition and convert the new model into a small-scale L2,p norm minimization problem, which requires no SVD and thus has low computational cost. An efficient algorithm based on alternating direction method is designed to solve the proposed problem. Experimental results on both synthetic and real-world data sets demonstrate the superiority of the proposed method over the state-of-the-art methods.	en_US
dc.relation.ispartof	IEEE Access	en_US
dc.relation.isbasedon	10.1109/ACCESS.2017.2765688	en_US
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	A Fast Generalized Low Rank Representation Framework Based on L<inf>2,p</inf> Norm Minimization for Subspace Clustering	en_US
dc.type	Journal Article
utslib.citation.volume	5	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
utslib.for	08 Information and Computing Sciences	en_US
utslib.for	09 Engineering	en_US
utslib.for	10 Technology	en_US
pubs.embargo.period	Not known	en_US
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 Electrical and Data Engineering
utslib.copyright.status	closed_access	*
pubs.publication-status	Published	en_US
pubs.volume	5	en_US

Abstract:

© 2017 IEEE. Low rank representation (LRR) is powerful for subspace clustering due to its strong ability in exploring low-dimensional subspace structures embedded in data. LRR is usually solved by iterative nuclear norm minimization, which involves singular value decomposition (SVD) at each iteration. However, the multiple SVDs limit the application of LRR due to its high computational cost. In this paper, we propose fast generalized LRR to address the above issue. Specifically, the nuclear norm and L2,1 norm in LRR are generalized to be the Schatten-p norm and L2,q norm, respectively. The new model is more general and robust than LRR. Then, we decompose the data matrix by Qatar riyal decomposition and convert the new model into a small-scale L2,p norm minimization problem, which requires no SVD and thus has low computational cost. An efficient algorithm based on alternating direction method is designed to solve the proposed problem. Experimental results on both synthetic and real-world data sets demonstrate the superiority of the proposed method over the state-of-the-art methods.

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

http://hdl.handle.net/10453/126817