File(s) under permanent embargo
Lingo: Linearized Grassmannian optimization for nuclear norm minimization
conference contribution
posted on 2015-01-01, 00:00 authored by Q Li, W Niu, Gang LiGang Li, Y Cao, J Tan, L GuoAs a popular heuristic to the matrix rank minimization problem, nuclear norm minimization attracts intensive research attentions. Matrix factorization based algorithms can reduce the expensive computation cost of SVD for nuclear norm minimization. However, most matrix factorization based algorithms fail to provide the theoretical guarantee for convergence caused by their non-unique factorizations. This paper proposes an efficient and accurate Linearized Grass-mannian Optimization (Lingo) algorithm, which adopts matrix factorization and Grassmann manifold structure to alternatively minimize the subproblems. More specially, linearization strategy makes the auxiliary variables unnecessary and guarantees the close-form solution for low periteration complexity. Lingo then converts linearized objective function into a nuclear norm minimization over Grass-mannian manifold, which could remedy the non-unique of solution for the low-rank matrix factorization. Extensive comparison experiments demonstrate the accuracy and efficiency of Lingo algorithm. The global convergence of Lingo is guaranteed with theoretical proof, which also verifies the effectiveness of Lingo.
History
Event
ACM International Conference on Information and Knowledge Management (24th : 2015 : Melbourne, Victoria)Pagination
801 - 809Publisher
ACM: The Association for Computing MachineryLocation
Melbourne, VictoriaPlace of publication
New York, N.Y.Publisher DOI
Start date
2015-10-19End date
2015-10-23ISBN-13
9781450337946Language
engPublication classification
E Conference publication; E1 Full written paper - refereedCopyright notice
2015, The Association for Computing MachineryTitle of proceedings
CIKM 2015: Proceedings of the 24th ACM International Conference on Information and Knowledge ManagementUsage metrics
Categories
No categories selectedLicence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC