MLJun 7, 2012

Robust subspace recovery by Tyler's M-estimator

arXiv:1206.1386v45 citations
Originality Synthesis-oriented
AI Analysis

This addresses robust subspace estimation for data analysis, but it is incremental as it applies an existing estimator to a known problem.

The paper tackles robust subspace recovery by showing that Tyler's M-estimator can recover a d-dimensional subspace from N points in R^D when the inlier percentage exceeds d/D and points are in general position, with empirical results favoring it over other convex algorithms in simulations and real data.

This paper considers the problem of robust subspace recovery: given a set of $N$ points in $\mathbb{R}^D$, if many lie in a $d$-dimensional subspace, then can we recover the underlying subspace? We show that Tyler's M-estimator can be used to recover the underlying subspace, if the percentage of the inliers is larger than $d/D$ and the data points lie in general position. Empirically, Tyler's M-estimator compares favorably with other convex subspace recovery algorithms in both simulations and experiments on real data sets.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes