On convergence of the inexact Rayleigh quotient iteration with MINRES

DOI10.1016/J.CAM.2012.05.016zbMATH Open1254.65048arXiv0906.2238OpenAlexW1966082499MaRDI QIDQ442709FDOQ442709

Authors: Zhongxiao Jia 

Publication date: 3 August 2012

Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Abstract: For the Hermitian inexact Rayleigh quotient iteration (RQI), we present a new general theory, independent of iterative solvers for shifted inner linear systems. The theory shows that the method converges at least quadratically under a new condition, called the uniform positiveness condition, that may allow inner tolerance $x{i}_{k}geq1$ at outer iteration $k$ and can be considerably weaker than the condition $x{i}_{k}leqxi<1$ with $xi$ a constant not near one commonly used in literature. We consider the convergence of the inexact RQI with the unpreconditioned and tuned preconditioned MINRES method for the linear systems. Some attractive properties are derived for the residuals obtained by MINRES. Based on them and the new general theory, we make a more refined analysis and establish a number of new convergence results. Let $|r_k|$ be the residual norm of approximating eigenpair at outer iteration $k$. Then all the available cubic and quadratic convergence results require $x{i}_k=O(|r_k|)$ and $x{i}_{k}leqxi$ with a fixed $xi$ not near one, respectively. Fundamentally different from these, we prove that the inexact RQI with MINRES generally converges cubically, quadratically and linearly provided that $x{i}_{k}leqxi$ with a constant $xi<1$ not near one, $x{i}_k=1-O(|r_k|)$ and $x{i}_k=1-O(|r_k{|}^2)$, respectively. Therefore, the new convergence conditions are much more relaxed than ever before. The theory can be used to design practical stopping criteria to implement the method more effectively. Numerical experiments confirm our results.

Full work available at URL: https://arxiv.org/abs/0906.2238

Recommendations

• Convergence Analysis of Iterative Solvers in Inexact Rayleigh Quotient Iteration
• On convergence of the inexact Rayleigh quotient iteration with the Lanczos method used for solving linear systems
• Convergence analysis of an inexact truncated RQ-iteration
• Convergence Analysis of Inexact Rayleigh Quotient Iteration
• Efficient Preconditioned Inner Solves For Inexact Rayleigh Quotient Iteration And Their Connections To The Single-Vector Jacobi–Davidson Method

zbMATH Keywords

convergenceinexact RQIinner iteration

Mathematics Subject Classification ID

Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Iterative numerical methods for linear systems (65F10)

Cites Work

• Title not available (Why is that?)
• Matrix algorithms. Vol. 2: Eigensystems
• Title not available (Why is that?)
• Title not available (Why is that?)
• Approximate solutions and eigenvalue bounds from Krylov subspaces
• On convergence of the inexact Rayleigh quotient iteration with the Lanczos method used for solving linear systems
• Inexact Rayleigh quotient-type methods for eigenvalue computations
• Inexact inverse iteration for symmetric matrices
• Computational methods for large eigenvalue problems
• Convergence Analysis of Iterative Solvers in Inexact Rayleigh Quotient Iteration
• Convergence of inexact inverse iteration with application to preconditioned iterative solvers
• A convergence analysis of the inexact Rayleigh quotient iteration and simplified Jacobi-Davidson method for the large Hermitian matrix eigenproblem
• The effects of inexact solvers in algorithms for symmetric eigenvalue problems
• Efficient Preconditioned Inner Solves For Inexact Rayleigh Quotient Iteration And Their Connections To The Single-Vector Jacobi–Davidson Method
• A tuned preconditioner for inexact inverse iteration applied to Hermitian eigenvalue problems
• Convergence Analysis of Inexact Rayleigh Quotient Iteration
• The convergence of Jacobi–Davidson iterations for Hermitian eigenproblems
• Inexact Inverse Iteration with Variable Shift for Nonsymmetric Generalized Eigenvalue Problems

Cited In (11)

• Convergence Analysis of Inexact Rayleigh Quotient Iteration
• A positivity preserving inexact Noda iteration for computing the smallest eigenpair of a large irreducible \(M\)-matrix
• On convergence of iterative projection methods for symmetric eigenvalue problems
• Title not available (Why is that?)
• On convergence of MRQI and IMRQI methods for Hermitian eigenvalue problems
• Title not available (Why is that?)
• Convergence proof of the harmonic Ritz pairs of iterative projection methods with restart strategies for symmetric eigenvalue problems
• Convergence Analysis of Iterative Solvers in Inexact Rayleigh Quotient Iteration
• On convergence of the inexact Rayleigh quotient iteration with the Lanczos method used for solving linear systems
• Inexact Rayleigh quotient-type methods for eigenvalue computations
• Tuned preconditioners for inexact two-sided inverse and Rayleigh quotient iteration.

Uses Software

• JDCG

This page was built for publication: On convergence of the inexact Rayleigh quotient iteration with MINRES

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q442709)