Abstract:We study the problem of determining if an input matrix A εRm x n can be well-approximated by a low rank matrix. Specifically, we study the problem of quickly estimating the rank or stable rank of A, the latter often providing a more robust measure of the rank. Since we seek significantly sublinear time algorithms, we cast these problems in the property testing framework. In this framework, A either has low rank or stable rank, or is far from having this property. The algorithm should read only a small number of entries or rows of A and decide which case A is in with high probability. If neither case occurs, the output is allowed to be arbitrary. We consider two notions of being far: (1) A requires changing at least an ε-fraction of its entries, or (2) A requires changing at least an ε-fraction of its rows. We call the former the "entry model" and the latter the "row model". We show: For testing if a matrix has rank at most d in the entry model, we improve the previous number of entries of A that need to be read from O(d2/ε2) (Krauthgamer and Sasson, SODA 2003) to O(d2/ε). Our algorithm is the first to adaptively query the entries of A, which for constant d we show is necessary to achieve O(1/ε) queries. For the important case of d = 1 we also give a new non-adaptive algorithm, improving the previous O(1/ε2) queries to O(log2(1/ε) / ε). For testing if a matrix has rank at most d in the row model, we prove an Ω(d/ε) lower bound on the number of rows that need to be read, even for adaptive algorithms. Our lower bound matches a non-adaptive upper bound of Krauthgamer and Sasson. For testing if a matrix has stable rank at most d in the row model or requires changing an ε/d-fraction of its rows in order to have stable rank at most d, we prove that reading θ(d/ε2) rows is necessary and sufficient. We also give an empirical evaluation of our rank and stable rank algorithms on real and synthetic datasets. For testing if a matrix has rank at most d in the entry model, we improve the previous number of entries of A that need to be read from O(d2/ε2) (Krauthgamer and Sasson, SODA 2003) to O(d2/ε). Our algorithm is the first to adaptively query the entries of A, which for constant d we show is necessary to achieve O(1/ε) queries. For the important case of d = 1 we also give a new non-adaptive algorithm, improving the previous O(1/ε2) queries to O(log2(1/ε) / ε). For testing if a matrix has rank at most d in the row model, we prove an Ω(d/ε) lower bound on the number of rows that need to be read, even for adaptive algorithms. Our lower bound matches a non-adaptive upper bound of Krauthgamer and Sasson. For testing if a matrix has stable rank at most d in the row model or requires changing an ε/d-fraction of its rows in order to have stable rank at most d, we prove that reading θ(d/ε2) rows is necessary and sufficient.

Relationship between the characteristic polynomial and the spectrum of a diagonalizable matrix and those of its low-rank update

Low-Rank Updates of Matrix Square Roots

Novel Alternating Update Method for Low Rank Approximation of Structured Matrices

Efficiently updating a covariance matrix and its LDL decomposition

Smoothed analysis of the condition number under low-rank perturbations

A general error analysis for randomized low-rank approximation with application to data assimilation

Multiple-rank modification of symmetric eigenvalue problem

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

Adaptive iterative singular value thresholding algorithm to low-rank matrix recovery

On low-rank approximability of solutions to high-dimensional operator equations and eigenvalue problems

Improved testing of low rank matrices

A Sherman--Morrison--Woodbury approach to solving least squares problems with low-rank updates

Randomized algorithms for low-rank matrix approximation: Design, analysis, and applications

Learning-Based Low-Rank Approximations

Randomized methods for rank-deficient linear systems

On the Asymptotic Nonnegative Rank of Matrices and its Applications in Information Theory

Matrix Nearness Problems with Off-Block-Diagonal Rank Constraints

A Simple Yet Efficient Rank One Update for Covariance Matrix Adaptation.

Adaptive Low-Rank Methods: Problems on Sobolev Spaces

Exploiting the Structure Effectively and Efficiently in Low Rank Matrix Recovery.

Low-Rank Kernel Matrix Approximation Using Skeletonized Interpolation With Endo- or Exo-Vertices