A geometric analysis of subspace clustering with outliers

TL;DR

This paper provides a geometric analysis of sparse subspace clustering (SSC), demonstrating its effectiveness in identifying multiple, intersecting subspaces with outliers, without prior knowledge of subspace count or dimensions.

Contribution

The paper introduces a novel geometric framework for analyzing SSC, extending its applicability to intersecting subspaces and datasets with many outliers, with theoretical guarantees.

Findings

SSC can recover multiple subspaces of high dimension

SSC correctly clusters data with intersecting subspaces

Extended SSC handles datasets with many outliers

Abstract

This paper considers the problem of clustering a collection of unlabeled data points assumed to lie near a union of lower-dimensional planes. As is common in computer vision or unsupervised learning applications, we do not know in advance how many subspaces there are nor do we have any information about their dimensions. We develop a novel geometric analysis of an algorithm named sparse subspace clustering (SSC) [In IEEE Conference on Computer Vision and Pattern Recognition, 2009. CVPR 2009 (2009) 2790-2797. IEEE], which significantly broadens the range of problems where it is provably effective. For instance, we show that SSC can recover multiple subspaces, each of dimension comparable to the ambient dimension. We also prove that SSC can correctly cluster data points even when the subspaces of interest intersect. Further, we develop an extension of SSC that succeeds when the data set is corrupted with possibly overwhelmingly many outliers. Underlying our analysis are clear geometric insights, which may bear on other sparse recovery problems. A numerical study complements our theoretical analysis and demonstrates the effectiveness of these methods.