Abstract:In this work, we consider the matrix completion problem, where the objective is to reconstruct a low-rank matrix from a few observed entries. A commonly employed approach involves nuclear norm minimization. For this method to succeed, the number of observed entries needs to scale at least proportional to both the rank of the ground-truth matrix and the coherence parameter. While the only prior information is oftentimes the low-rank nature of the ground-truth matrix, in various real-world scenarios, additional knowledge about the ground-truth low-rank matrix is available. For instance, in collaborative filtering, Netflix problem, and dynamic channel estimation in wireless communications, we have partial or full knowledge about the signal subspace in advance. Specifically, we are aware of some subspaces that form multiple angles with the column and row spaces of the ground-truth matrix. Leveraging this valuable information has the potential to significantly reduce the required number of observations. To this end, we introduce a multi-weight nuclear norm optimization problem that concurrently promotes the low-rank property as well the information about the available subspaces. The proposed weights are tailored to penalize each angle corresponding to each basis of the prior subspace independently. We further propose an optimal weight selection strategy by minimizing the coherence parameter of the ground-truth matrix, which is equivalent to minimizing the required number of observations. Simulation results validate the advantages of incorporating multiple weights in the completion procedure. Specifically, our proposed multi-weight optimization problem demonstrates a substantial reduction in the required number of observations compared to the state-of-the-art methods.

Sources Localization Through Matrix Completion Via Nystrom Completion

Squared Distance Matrix Completion Through Nystrom Approximation.

High Resolution Matched-Field Source Localization Based on Sparse-Reconstruction

A High Resolution Bearing Estimation Method Based on Sparse Reconstruction

Low Complexity Modeling of Cross-Spectral Matrix and Its Application in the Non-Synchronous Measurements of Microphones Array

Sources Localization Via Multi-Dimesional Scaling Based on Subspace Method

Structural-Missing Tensor Completion for Robust DOA Estimation with Sensor Failure

Doa Estimation Via Coarray Tensor Completion with Missing Slices.

Locally Linear Approximation Approach for Incomplete Data.

Location of SMOS RFI Sources Using a Matrix Completion Approach

Matrix Completion with Nonuniform Sampling: Theories and Methods.

Localization from Incomplete Euclidean Distance Matrix: Performance Analysis for the SVD-MDS Approach

Subspace projection matrix recovery from incomplete information

Near-Field Multiple Sources Localization Via Sparse Reconstruction

Calibration Using Matrix Completion with Application to Ultrasound Tomography

Unimodality-Constrained Matrix Factorization for Non-Parametric Source Localization

Low Multilinear Rank Tensor Completion with Missing Data

Matrix Completion with Deterministic Sampling: Theories and Methods

Subspace-Informed Matrix Completion

A Matrix Completion Based Method for RFI Source Localization in Microwave Interferometric Radiometry

Seismic Data Reconstruction Via Matrix Completion