Abstract:Recovering a large matrix from a small subset of its entries is a challenging problem arising in many real applications, such as image inpainting and recommender systems. Many existing approaches formulate this problem as a general low-rank matrix approximation problem. Since the rank operator is nonconvex and discontinuous, most of the recent theoretical studies use the nuclear norm as a convex relaxation. One major limitation of the existing approaches based on nuclear norm minimization is that all the singular values are simultaneously minimized, and thus the rank may not be well approximated in practice. In this paper, we propose to achieve a better approximation to the rank of matrix by truncated nuclear norm, which is given by the nuclear norm subtracted by the sum of the largest few singular values. In addition, we develop a novel matrix completion algorithm by minimizing the Truncated Nuclear Norm. We further develop three efficient iterative procedures, TNNR-ADMM, TNNR-APGL, and TNNR-ADMMAP, to solve the optimization problem. TNNR-ADMM utilizes the alternating direction method of multipliers (ADMM), while TNNR-AGPL applies the accelerated proximal gradient line search method (APGL) for the final optimization. For TNNR-ADMMAP, we make use of an adaptive penalty according to a novel update rule for ADMM to achieve a faster convergence rate. Our empirical study shows encouraging results of the proposed algorithms in comparison to the state-of-the-art matrix completion algorithms on both synthetic and real visual datasets.

Efficient Sparse Low-Rank Tensor Completion Using the Frank-Wolfe Algorithm.

An Efficient Tensor Completion Method Via New Latent Nuclear Norm

Tensor Completion Via a Multi-Linear Low-N-rank Factorization Model

Low rank tensor completion with sparse regularization in a transformed domain

Beyond Low Rank: A Data-Adaptive Tensor Completion Method

Exploiting Global Low-Rank Structure and Local Sparsity Nature for Tensor Completion.

Efficient enhancement of low-rank tensor completion via thin QR decomposition

Fast and Accurate Matrix Completion Via Truncated Nuclear Norm Regularization

Low-Rank Tensor Completion via Novel Sparsity-Inducing Regularizers

Tensor Factorization for Low-Rank Tensor Completion.

Logarithmic Norm Regularized Low-Rank Factorization for Matrix and Tensor Completion

Scalable Tensor Completion with Nonconvex Regularization.

Low-Rank Tensor Completion Based on Log-Det Rank Approximation and Matrix Factorization

A Low-Rank Tensor Completion Method via Strassen–Ottaviani Flattening

FasTer: Fast Tensor Completion with Nonconvex Regularization

Imbalanced low-rank tensor completion via latent matrix factorization

A dual framework for low-rank tensor completion

A Scale-Invariant Relaxation in Low-Rank Tensor Recovery with an Application to Tensor Completion

Efficient Nonconvex Regularized Tensor Completion with Structure-aware Proximal Iterations

Accurate Tensor Completion via Adaptive Low-Rank Representation

Solving Low-Rank Matrix Completion Problems Efficiently