Robust Matrix Decomposition with Outliers

Abstract

Suppose a given observation matrix can be decomposed as the sum of a low-rank matrix and a sparse matrix (outliers), and the goal is to recover these individual components from the observed sum. Such additive decompositions have applications in a variety of numerical problems including system identification, latent variable graphical modeling, and principal components analysis. We study conditions under which recovering such a decomposition is possible via a combination of ${\ell }_1$ norm and trace norm minimization. We are specifically interested in the question of how many outliers are allowed so that convex programming can still achieve accurate recovery, and we obtain stronger recovery guarantees than previous studies. Moreover, we do not assume that the spatial pattern of outliers is random, which stands in contrast to related analyses under such assumptions via matrix completion.

Corrected comparisons to previous work of Candes et al (2009)

References in corpus (1)

• Latent variable graphical model selection via convex optimization

Cited by in corpus (8)

• An Online Algorithm for Separating Sparse and Low-dimensional Signal Sequences from their Sum
• Low-Rank Structure Learning via Log-Sum Heuristic Recovery
• Compressive Principal Component Pursuit
• Low-rank Matrix Recovery from Errors and Erasures
• Fast global convergence of gradient methods for high-dimensional statistical recovery
• Square Root Principal Component Pursuit: Tuning-Free Noisy Robust Matrix Recovery
• High-dimensional Joint Sparsity Random Effects Model for Multi-task Learning
• Recursive Sparse Recovery in Large but Structured Noise - Part 2