2013/01/31 by Mahdi Soltanolkotabi, Ehsan Elhamifar, Emmanuel J. Candès · 1 citation
Computer Science · Engineering · Mathematics · #Advanced Clustering Algorithms Research #Face and Expression Recognition #Sparse and Compressive Sensing Techniques #cs.IT #cs.LG #math.IT #math.OC #math.ST #stat.ML #stat.TH
paper · pdf · doi:10.1214/13-aos1199
published as Annals of Statistics 2014, Vol. 42, No. 2, 669-699 · Published in at http://dx.doi.org/10.1214/13-AOS1199 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
openalex publication_date 2014/04/01 · arxiv created 2014/05/23 · arxiv updated 2014/05/26 · openalex created_date 2016/06/24 · openalex updated_date 2026/07/28
Subspace clustering refers to the task of finding a multi-subspace representation that best fits a collection of points taken from a high-dimensional space. This paper introduces an algorithm inspired by sparse subspace clustering (SSC) [In IEEE Conference on Computer Vision and Pattern Recognition, CVPR (2009) 2790–2797] to cluster noisy data, and develops some novel theory demonstrating its correctness. In particular, the theory uses ideas from geometric functional analysis to show that the algorithm can accurately recover the underlying subspaces under minimal requirements on their orientation, and on the number of samples per subspace. Synthetic as well as real data experiments complement our theoretical study, illustrating our approach and demonstrating its effectiveness.