Low-Rank Matrix Approximations Do Not Need a Singular Value Gap

DOI10.1137/18M1163658zbMath1455.65066arXiv1801.00670OpenAlexW2962811996WikidataQ128382729 ScholiaQ128382729MaRDI QIDQ3119542

Ilse C. F. Ipsen, Petros Drineas

Publication date: 12 March 2019

Published in: SIAM Journal on Matrix Analysis and Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1801.00670

zbMATH Keywords

singular value decompositionmultiplicative perturbationsprincipal anglesadditive perturbations

Mathematics Subject Classification ID

Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Inequalities involving eigenvalues and eigenvectors (15A42) Eigenvalues, singular values, and eigenvectors (15A18) Numerical methods for low-rank matrix approximation; matrix compression (65F55)

Related Items

Perturbations of the \textsc{Tcur} decomposition for tensor valued data in the Tucker format, A Theory of Quantum Subspace Diagonalization, A unified performance analysis of likelihood-informed subspace methods, Admissible subspaces and the subspace iteration method, Randomized Subspace Iteration: Analysis of Canonical Angles and Unitarily Invariant Norms, Randomized block Krylov methods for approximating extreme eigenvalues, Unnamed Item, Perturbations of CUR Decompositions, Sensitivity of low-rank matrix recovery

Uses Software

• mctoolbox
• RandNLA

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• History and generality of the CS decomposition
• An overview of relative \(\sin\Theta\) theorems for invariant subspaces of complex matrices
• A note on element-wise matrix sparsification via a matrix-valued Bernstein inequality
• New estimates for the recursive low-rank truncation of block-structured matrices
• Computational Advertising: Techniques for Targeting Relevant Ads
• Numerical Methods for Large Eigenvalue Problems
• On the Numerical Analysis of Oblique Projectors
• Fast computation of low-rank matrix approximations
• On the Rates of Convergence of the Lanczos and the Block-Lanczos Methods
• Error and Perturbation Bounds for Subspaces Associated with Certain Eigenvalue Problems
• Accuracy and Stability of Numerical Algorithms
• Randomized Approximation of the Gram Matrix: Exact Computation and Probabilistic Bounds
• Subspace Iteration Randomization and Singular Value Problems
• Structural Convergence Results for Approximation of Dominant Subspaces from Block Krylov Spaces
• Fast Monte Carlo Algorithms for Matrices I: Approximating Matrix Multiplication
• Fast Monte Carlo Algorithms for Matrices II: Computing a Low-Rank Approximation to a Matrix
• Some new bounds on perturbation of subspaces
• The Rotation of Eigenvectors by a Perturbation. III
• Perturbation bounds in connection with singular value decomposition
• Angles between subspaces and their tangents