Abstract:In this paper, we study the theoretical properties of iteratively reweighted least squares algorithm for recovering a matrix (IRLS-M for short) from noisy linear measurements. The IRLS-M was proposed by Fornasier et al. (2011) [ 17 ] for solving nuclear norm minimization and by Mohan et al. (2012) [ 31 ] for solving Schatten- \begin{document}$p$\end{document} (quasi) norm minimization ( \begin{document}$0 ) in noiseless case, based on the iteratively reweighted least squares algorithm for sparse signal recovery (IRLS for short) (Daubechies et al., 2010) [ 15 ], and numerical experiments have been given to show its efficiency (Fornasier et al. and Mohan et al.) [ 17 ], [ 31 ]. In this paper, we focus on providing convergence and stability analysis of iteratively reweighted least squares algorithm for low-rank matrix recovery in the presence of noise. The convergence of IRLS-M is proved strictly for all \begin{document}$0 . Furthermore, when the measurement map \begin{document}$\mathcal{A}$\end{document} satisfies the matrix restricted isometry property (M-RIP for short), we show that the IRLS-M is stable for \begin{document}$0 . Specially, when \begin{document}$p=1$\end{document} , we prove that the M-RIP constant \begin{document}$δ_{2r} is sufficient for IRLS-M to recover an unknown (approximately) low rank matrix with an error that is proportional to the noise level. The simplicity of IRLS-M, along with the theoretical guarantees provided in this paper, make a compelling case for its adoption as a standard tool for low rank matrix recovery.

Reweighted Nonnegative Least-Mean-Square Algorithm

CONVERGENCE AND STABILITY OF ITERATIVELY REWEIGHTED LEAST SQUARES FOR LOW-RANK MATRIXRECOVERY

Kernel Least-Mean Mixed-Norm Algorithm

A Fast Scale-Invariant Algorithm for Non-negative Least Squares with Non-negative Data

A bias-compensated NLMS algorithm based on arctangent framework for system identification

Sparse Normalized Least Mean Absolute Deviation Algorithm Based on Unbiasedness Criterion for System Identification with Noisy Input

l0 Norm Constraint LMS Algorithm for Sparse System Identification.

A Bias-Compensated Fractional Order Normalized Least Mean Square Algorithm with Noisy Inputs

Robust Bias Compensation Method for Sparse Normalized Quasi-Newton Least-Mean with Variable Mixing-Norm Adaptive Filtering

Nonnegative Low Rank Matrix Approximation for Nonnegative Matrices

An efficient non-negative least mean squares algorithm based on q-gradient for system identification

Regularized Least-Mean-Square Algorithms

Correntropy induced metric penalized NLMF algorithm to improve sparse system identification

A Gradient-Based Algorithm with Nonmonotone Line Search for Nonnegative Matrix Factorization

Recursive Direct Weight Optimization in Nonlinear System Identification: A Minimal Probability Approach

Precise asymptotics of reweighted least-squares algorithms for linear diagonal networks

$L_{0}$ Norm Constraint LMS Algorithm for Sparse System Identification

Variable Step-Size NLMP Algorithm with a Gradient-Based Weighted Average in Impulsive Environments

Robust Zero-Point Attraction Least Mean Square Algorithm on Near Sparse System Identification

Noise constrained least mean absolute third algorithm

A Minimal Probability Approach in Nonparametric Nonlinear System Identification