Abstract:This paper proposes scalable and fast algorithms for solving the Robust PCA problem, namely recovering a low-rank matrix with an unknown fraction of its entries being arbitrarily corrupted. This problem arises in many applications, such as image processing, web data ranking, and bioinformatic data analysis. It was recently shown that under surprisingly broad conditions, the Robust PCA problem can be exactly solved via convex optimization that minimizes a combination of the nuclear norm and the ℓ^1-norm . In this paper, we apply the method of augmented Lagrange multipliers (ALM) to solve this convex program. As the objective function is non-smooth, we show how to extend the classical analysis of ALM to such new objective functions and prove the optimality of the proposed algorithms and characterize their convergence rate. Empirically, the proposed new algorithms can be more than five times faster than the previous state-of-the-art algorithms for Robust PCA, such as the accelerated proximal gradient (APG) algorithm. Moreover, the new algorithms achieve higher precision, yet being less storage/memory demanding. We also show that the ALM technique can be used to solve the (related but somewhat simpler) matrix completion problem and obtain rather promising results too. We further prove the necessary and sufficient condition for the inexact ALM to converge globally. Matlab code of all algorithms discussed are available at http://perception.csl.illinois.edu/matrix-rank/home.html

Nonconvex relaxation approaches to robust matrix recovery

Robust Recovery of Low-Rank Matrices Via Non-Convex Optimization

Robust Sparse Recovery Via Non-Convex Optimization

Nonconvex Robust Low-rank Matrix Recovery

A NEW MODEL FOR SPARSE AND LOW-RANK MATRIX DECOMPOSITION

Low-rank matrix recovery via nonconvex optimization methods with application to errors-in-variables matrix regression

A multi-stage convex relaxation approach to noisy structured low-rank matrix recovery

Robust PCA Using Generalized Nonconvex Regularization.

Proximity Point Algorithm for Low-Rank Matrix Recovery from Sparse Noise Corrupted Data

The Perturbation Analysis of Nonconvex Low-Rank Matrix Robust Recovery

Robust Low-Rank Matrix Recovery As Mixed Integer Programming Via $\ell _{0}$-Norm Optimization.

A non-convex approach to low-rank and sparse matrix decomposition.

Pathwise Component Descent Method with MC+ Penalty for Low Rank Matrix Recovery

Simultaneously Sparse and Low-Rank Matrix Reconstruction Via Nonconvex and Nonseparable Regularization

Efficient Robust Matrix Factorization with Nonconvex Penalties.

A Non-convex Approach for Sparse Recovery with Convergence Guarantee

Improved Nonconvex Optimization Model for Low-Rank Matrix Recovery

Low-rank Matrix Recovery Via Novel Double Nonconvex Nonsmooth Rank Minimization with ADMM

Low-rank matrix recovery with non-quadratic loss: projected gradient method and regularity projection oracle

The Augmented Lagrange Multiplier Method for Exact Recovery of Corrupted Low-Rank Matrices

A New Nonconvex Rank Approximation of RPCA