Optimal rank-1 Hankel approximation of matrices: Frobenius norm and spectral norm and Cadzow's algorithm

DOI10.1016/J.LAA.2021.07.004zbMATH Open1472.15041arXiv2004.11099OpenAlexW3185453068MaRDI QIDQ1979358FDOQ1979358

Authors: Hanna Knirsch, Markus Petz, Gerlind Plonka 

Publication date: 2 September 2021

Published in: Linear Algebra and its Applications (Search for Journal in Brave)

Abstract: We characterize optimal rank-1 matrix approximations with Hankel or Toeplitz structure with regard to two different norms, the Frobenius norm and the spectral norm, in a new way. More precisely, we show that these rank-1 matrix approximation problems can be solved by maximizing special rational functions. Our approach enables us to show that the optimal solutions with respect to these two norms have completely different structure and only coincide in the trivial case when the singular value decomposition already provides an optimal rank-1 approximation with the desired Hankel or Toeplitz structure. We also prove that the Cadzow algorithm for structured low-rank approximations always converges to a fixed point in the rank-1 case. However, it usually does not converge to the optimal solution, neither with regard to the Frobenius norm nor the spectral norm.

Full work available at URL: https://arxiv.org/abs/2004.11099

Recommendations

• scientific article; zbMATH DE number 1086779
• Low rank approximation of a Hankel matrix by structured total least norm
• Block-row Hankel weighted low rank approximation
• A gradient system approach for Hankel structured low-rank approximation
• scientific article; zbMATH DE number 5004921

zbMATH Keywords

spectral normHankel matricesToeplitz matricesFrobenius normCadzow algorithmoptimal structured low-rank approximation

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Canonical forms, reductions, classification (15A21) Norms of matrices, numerical range, applications of functional analysis to matrix theory (15A60) Toeplitz, Cauchy, and related matrices (15B05) Linear operator approximation theory (47A58) System structure simplification (93B11)

Cites Work

• Matrix Analysis
• Hankel matrix rank minimization with applications to system identification and realization
• Exact solutions in structured low-rank approximation
• On approximation of functions by exponential sums
• Title not available (Why is that?)
• ANALYTIC PROPERTIES OF SCHMIDT PAIRS FOR A HANKEL OPERATOR AND THE GENERALIZED SCHUR-TAKAGI PROBLEM
• Algebraic methods for Toeplitz-like matrices and operators
• On approximate GCDs of univariate polynomials
• Structured low rank approximation
• Cadzow denoising upgraded: a new projection method for the recovery of Dirac pulses from noisy linear measurements
• Interior-point method for nuclear norm approximation with application to system identification
• Signal enhancement-a composite property mapping algorithm
• Prony methods for recovery of structured functions
• Alternating Projections on Manifolds
• Sampling signals with finite rate of innovation
• Variable projection for affinely structured low-rank approximation in weighted \(2\)-norms
• Title not available (Why is that?)
• A Modified Prony Algorithm for Exponential Function Fitting
• Alternating projections on nontangential manifolds
• Structured Perturbations Part I: Normwise Distances
• A subdivision method for computing nearest gcd with certification
• Title not available (Why is that?)
• Analysis of structured low rank approximation as an optimization problem
• Fast algorithm for solving the Hankel/Toeplitz structured total least squares problem
• On the choice of parameters in singular spectrum analysis and related subspace-based methods
• Structured Perturbations Part II: Componentwise Distances
• Application of structured total least squares for system identification and model reduction
• Structured total least squares and \(L_ 2\) approximation problems
• Fixed-point algorithms for frequency estimation and structured low rank approximation
• Analysis of the structured total least squares problem for Hankel/Toeplitz matrices
• Computation of adaptive Fourier series by sparse approximation of exponential sums
• Factorization approach to structured low-rank approximation with applications
• Low-Rank Optimization With Convex Constraints
• Iterative algorithms for weighted and unweighted finite-rank time-series approximations
• Optimal approximation with exponential sums by a maximum likelihood modification of Prony's method
• The principal minor test for semidefinite matrices

Cited In (1)

• Critical points and error rank in best \(H_2\) matrix rational approximation of fixed McMillan degree

This page was built for publication: Optimal rank-1 Hankel approximation of matrices: Frobenius norm and spectral norm and Cadzow's algorithm

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1979358)