Robust subspace clustering

TL;DR

This paper presents a robust subspace clustering algorithm inspired by SSC, with theoretical guarantees for accurate recovery of subspaces from noisy high-dimensional data, supported by experiments on synthetic and real datasets.

Contribution

It introduces a new robust clustering algorithm with novel theoretical analysis ensuring correct subspace recovery under minimal conditions.

Findings

Algorithm accurately clusters noisy data.

Theoretical guarantees for subspace recovery.

Effective on both synthetic and real datasets.

Abstract

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.