On the subspace projected approximate matrix method.

DOI10.1007/S10492-015-0104-8zbMATH Open1363.65048arXiv1103.1779OpenAlexW77078096MaRDI QIDQ499032FDOQ499032

Ricardo Reis da Silva, J. H. Brandts

Publication date: 29 September 2015

Published in: Applications of Mathematics (Search for Journal in Brave)

Abstract: We provide a comparative study of the Subspace Projected Approximate Matrix method, abbreviated SPAM, which is a fairly recent iterative method to compute a few eigenvalues of a Hermitian matrix $A$. It falls in the category of inner-outer iteration methods and aims to save on the costs of matrix-vector products with $A$ within its inner iteration. This is done by choosing an approximation $A_0$ of $A$, and then, based on both $A$ and $A_0$, to define a sequence $(A_k{)}_{k=0}^n$ of matrices that increasingly better approximate $A$ as the process progresses. Then the matrix $A_k$ is used in the $k$th inner iteration instead of $A$. In spite of its main idea being refreshingly new and interesting, SPAM has not yet been studied in detail by the numerical linear algebra community. We would like to change this by explaining the method, and to show that for certain special choices for $A_0$, SPAM turns out to be mathematically equivalent to known eigenvalue methods. More sophisticated approximations $A_0$ turn SPAM into a boosted version of Lanczos, whereas it can also be interpreted as an attempt to enhance a certain instance of the preconditioned Jacobi-Davidson method. Numerical experiments are performed that are specifically tailored to illustrate certain aspects of SPAM and its variations. For experiments that test the practical performance of SPAM in comparison with other methods, we refer to other sources. The main conclusion is that SPAM provides a natural transition between the Lanczos method and one-step preconditioned Jacobi-Davidson.

Full work available at URL: https://arxiv.org/abs/1103.1779

zbMATH Keywords

eigenvaluesnumerical experimentsJacobi-Davidson methodLanczos methodHermitian eigenproblemRitz-Galerkin approximationsubspace projected approximate matrix

Mathematics Subject Classification ID

Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Iterative numerical methods for linear systems (65F10) Numerical computation of matrix norms, conditioning, scaling (65F35)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Implicit Application of Polynomial Filters in a k-Step Arnoldi Method
• Templates for the Solution of Algebraic Eigenvalue Problems
• Matrix algorithms. Vol. 2: Eigensystems
• A Jacobi–Davidson Iteration Method for Linear Eigenvalue Problems
• Finite element methods (Part 1)
• Solution of Sparse Indefinite Systems of Linear Equations
• The Riccati algorithm for eigenvalues and invariant subspaces of matrices with inexpensive action
• A Krylov--Schur algorithm for large eigenproblems
• A Jacobi--Davidson Iteration Method for Linear Eigenvalue Problems
• The iterative calculation of a few of the lowest eigenvalues and corresponding eigenvectors of large real-symmetric matrices
• Generalizations of Davidson’s Method for Computing Eigenvalues of Sparse Symmetric Matrices
• An analysis of the Rayleigh-Ritz method for approximating eigenspaces
• The Davidson Method
• Eigenvalue computation in the 20th century
• Norms and exclusion theorems
• Computing eigenvalue bounds for iterative subspace matrix methods
• A Subspace-Projected Approximate Matrix Method for Systems of Linear Equations
• Harmonic Ritz and Lehmann bounds
• On the use of harmonic Ritz pairs in approximating internal eigenpairs
• Alternative correction equations in the Jacobi-Davidson method
• The subspace projected approximate matrix (SPAM) modification of the Davidson method
• Parallel implementation of efficient preconditioned linear solver for grid-based applications in chemical physics. I: Block Jacobi diagonalization

Cited In (5)

• Subspace projection variants on Newton's method
• Subspace Sampling and Relative-Error Matrix Approximation: Column-Based Methods
• Title not available (Why is that?)
• The subspace projected approximate matrix (SPAM) modification of the Davidson method
• Subfactor projections.

Uses Software

• eigs
• JDQR
• JDQZ
• IRAM

This page was built for publication: On the subspace projected approximate matrix method.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q499032)