arXiv Analytics

Sign in

arXiv:1202.3772 [cs.LG]AbstractReferencesReviewsResources

Rank/Norm Regularization with Closed-Form Solutions: Application to Subspace Clustering

Yao-Liang Yu, Dale Schuurmans

Published 2012-02-14, updated 2012-10-09Version 2

When data is sampled from an unknown subspace, principal component analysis (PCA) provides an effective way to estimate the subspace and hence reduce the dimension of the data. At the heart of PCA is the Eckart-Young-Mirsky theorem, which characterizes the best rank k approximation of a matrix. In this paper, we prove a generalization of the Eckart-Young-Mirsky theorem under all unitarily invariant norms. Using this result, we obtain closed-form solutions for a set of rank/norm regularized problems, and derive closed-form solutions for a general class of subspace clustering problems (where data is modelled by unions of unknown subspaces). From these results we obtain new theoretical insights and promising experimental results.

Comments: 11 pages, 1 figure, appeared in UAI 2011. One footnote corrected and appendix added
Categories: cs.LG, cs.NA, stat.ML
Related articles: Most relevant | Search more
arXiv:1804.05090 [cs.LG] (Published 2018-04-13)
Regularized Singular Value Decomposition and Application to Recommender System
arXiv:1408.4576 [cs.LG] (Published 2014-08-20)
Introduction to Clustering Algorithms and Applications
arXiv:1204.5309 [cs.LG] (Published 2012-04-24, updated 2013-03-26)
Analysis Operator Learning and Its Application to Image Reconstruction