Abstract:This paper introduces a novel algorithm to approximate the matrix with minimum nuclear norm among all matrices obeying a set of convex constraints. This problem may be understood as the convex relaxation of a rank minimization problem and arises in many important applications as in the task of recovering a large matrix from a small subset of its entries (the famous Netflix problem). Off-the-shelf algorithms such as interior point methods are not directly amenable to large problems of this kind with over a million unknown entries. This paper develops a simple first-order and easy-to-implement algorithm that is extremely efficient at addressing problems in which the optimal solution has low rank. The algorithm is iterative, produces a sequence of matrices $\{\boldsymbol{X}^k,\boldsymbol{Y}^k\}$, and at each step mainly performs a soft-thresholding operation on the singular values of the matrix $\boldsymbol{Y}^k$. There are two remarkable features making this attractive for low-rank matrix completion problems. The first is that the soft-thresholding operation is applied to a sparse matrix; the second is that the rank of the iterates $\{\boldsymbol{X}^k\}$ is empirically nondecreasing. Both these facts allow the algorithm to make use of very minimal storage space and keep the computational cost of each iteration low. On the theoretical side, we provide a convergence analysis showing that the sequence of iterates converges. On the practical side, we provide numerical examples in which $1,000\times1,000$ matrices are recovered in less than a minute on a modest desktop computer. We also demonstrate that our approach is amenable to very large scale problems by recovering matrices of rank about 10 with nearly a billion unknowns from just about 0.4% of their sampled entries. Our methods are connected with the recent literature on linearized Bregman iterations for $\ell_1$ minimization, and we develop a framework in which one can understand these algorithms in terms of well-known Lagrange multiplier algorithms.

An Upper Bound on the Minimum Rank of a Symmetric Toeplitz Matrix Completion Problem

The Bounds of Restricted Isometry Constants for Low Rank Matrices Recovery

On the Schatten P-Quasi-norm Minimization for Low-Rank Matrix Recovery

On the equivalence between low rank matrix completion and tensor rank

Upper and lower bounds for the rank of the matrix-valued function when X has a fixed rank

Exploring Algorithmic Limits of Matrix Rank Minimization under Affine Constraints

An approximation theory of matrix rank minimization and its application to quadratic equations

Optimal Low-Rank Matrix Completion: Semidefinite Relaxations and Eigenvector Disjunctions

The lower bounds for the rank of matrices and some sufficient conditions for nonsingular matrices

A Singular Value Thresholding Algorithm for Matrix Completion

Minimizing low-rank models of high-order tensors: Hardness, span, tight relaxation, and applications

Effective Lower Bounds on the Matrix Rank and Their Applications

Hybrid Singular Value Thresholding For Tensor Completion

Truncated quadratic norm minimization for bilinear factorization based matrix completion

Lower Bounds on the Mean-Squared Error of Low-Rank Matrix Reconstruction

Quasi-Hankel low-rank matrix completion: a convex relaxation

The Rank-1 Completion Problem for Cubic Tensors

Low Rank Matrix Completion via Robust Alternating Minimization in Nearly Linear Time

Solving Low-Rank Matrix Completion Problems Efficiently

Concise tensors of minimal border rank