FEAST as a subspace iteration eigensolver accelerated by approximate spectral projection
From MaRDI portal
Publication:2923348
DOI10.1137/13090866XzbMATH Open1303.65018arXiv1302.0432OpenAlexW2964332084MaRDI QIDQ2923348FDOQ2923348
Authors: Ping Tak Peter Tang, Eric Polizzi 
Publication date: 15 October 2014
Published in: SIAM Journal on Matrix Analysis and Applications (Search for Journal in Brave)
Abstract: The calculation of a segment of eigenvalues and their corresponding eigenvectors of a Hermitian matrix or matrix pencil has many applications. A new density-matrix-based algorithm has been proposed recently and a software package FEAST has been developed. The density-matrix approach allows FEAST's implementation to exploit a key strength of modern computer architectures, namely, multiple levels of parallelism. Consequently, the software package has been well received, especially in the electronic structure community. Nevertheless, theoretical analysis of FEAST has lagged. For instance, the FEAST algorithm has not been proven to converge. This paper offers a detailed numerical analysis of FEAST. In particular, we show that the FEAST algorithm can be understood as an accelerated subspace iteration algorithm in conjunction with the Rayleigh-Ritz procedure. The novelty of FEAST lies in its accelerator which is a rational matrix function that approximates the spectral projector onto the eigenspace in question. Analysis of the numerical nature of this approximate spectral projector and the resulting subspaces generated in the FEAST algorithm establishes the algorithm's convergence. This paper shows that FEAST is resilient against rounding errors and establishes properties that can be leveraged to enhance the algorithm's robustness. Finally, we propose an extension of FEAST to handle non-Hermitian problems and suggest some future research directions.
Full work available at URL: https://arxiv.org/abs/1302.0432
Recommendations
- Dissecting the FEAST algorithm for generalized eigenproblems
- Feast eigensolver for non-Hermitian problems
- A harmonic FEAST algorithm for non-Hermitian generalized eigenvalue problems
- On the non-Hermitian FEAST algorithms with oblique projection for eigenvalue problems
- Krylov eigenvalue strategy using the FEAST algorithm with inexact system solves.
Eigenvalues, singular values, and eigenvectors (15A18) Numerical computation of eigenvalues and eigenvectors of matrices (65F15)
Cited In (60)
- A subspace iteration eigensolver based on Cauchy integrals for vibroacoustic problems in unbounded domains
- A robust randomized indicator method for accurate symmetric eigenvalue detection
- An algorithm for identifying eigenvectors exhibiting strong spatial localization
- Complex moment-based eigensolver coupled with two Krylov subspaces
- An Augmented Matrix-Based CJ-FEAST SVDsolver for Computing a Partial Singular Value Decomposition with the Singular Values in a Given Interval
- A FEAST SVDsolver based on Chebyshev-Jackson series for computing partial singular triplets of large matrices
- A contour integral-based method for nonlinear eigenvalue problems for semi-infinite photonic crystals
- A rational filtering algorithm for sequences of shifted symmetric linear systems with applications to frequency response analysis
- On localization of eigenfunctions of the magnetic Laplacian
- A harmonic FEAST algorithm for non-Hermitian generalized eigenvalue problems
- A generalization of Saad's bound on harmonic Ritz vectors of Hermitian matrices
- Block Krylov-type complex moment-based eigensolvers for solving generalized eigenvalue problems
- Zolotarev quadrature rules and load balancing for the FEAST eigensolver
- Computing eigenvalues of real symmetric matrices with rational filters in real arithmetic
- Triangularized orthogonalization-free method for solving extreme eigenvalue problems
- A contour-integral based method with Schur-Rayleigh-Ritz procedure for generalized eigenvalue problems
- A novel numerical method to determine the algebraic multiplicity of nonlinear eigenvalues
- A contour-integral based method for counting the eigenvalues inside a region
- Spectral discretization errors in filtered subspace iteration
- Designing rational filter functions for solving eigenvalue problems by contour integration
- An efficient Gauss-Newton algorithm for symmetric low-rank product matrix approximations
- Randomized subspace iteration: analysis of canonical angles and unitarily invariant norms
- Analysis of FEAST spectral approximations using the DPG discretization
- FEAST for differential eigenvalue problems
- On Chebyshev-Davidson method for symmetric generalized eigenvalue problems
- On convergence of iterative projection methods for symmetric eigenvalue problems
- On the computation of large-scale self-consistent-field iterations
- Localized spectrum slicing
- Rational Spectral Filters with Optimal Convergence Rate
- Computing partial spectra with least-squares rational filters
- Bifurcation analysis of stationary solutions of two-dimensional coupled Gross-Pitaevskii equations using deflated continuation
- Multiscale modal analysis of fully-loaded spent nuclear fuel canisters
- Efficient estimation of eigenvalue counts in an interval.
- FEAST
- Fast Randomized Non-Hermitian Eigensolvers Based on Rational Filtering and Matrix Partitioning
- A subspace iteration algorithm for Fredholm valued functions
- Convergence of integration-based methods for the solution of standard and generalized Hermitian eigenvalue problems
- On the non-Hermitian FEAST algorithms with oblique projection for eigenvalue problems
- Increasing the performance of the Jacobi-Davidson method by blocking
- Twice is enough for dangerous eigenvalues
- A fast contour-integral eigensolver for non-Hermitian matrices
- Recursive integral method for transmission eigenvalues
- A spectral projection method for transmission eigenvalues
- Complex moment-based methods for differential eigenvalue problems
- A FEAST algorithm with oblique projection for generalized eigenvalue problems.
- Solving the three-dimensional high-frequency Helmholtz equation using contour integration and polynomial preconditioning
- Feast eigensolver for non-Hermitian problems
- Solving large-scale finite element nonlinear eigenvalue problems by resolvent sampling based Rayleigh-Ritz method
- Nonlinear eigenvalue problems and contour integrals
- Error bounds of Rayleigh-Ritz type contour integral-based eigensolver for solving generalized eigenvalue problems
- Relationships among contour integral-based methods for solving generalized eigenvalue problems
- Projection method for eigenvalue problems of linear nonsquare matrix pencils
- A numerical study of fractional linear algebraic systems
- Krylov eigenvalue strategy using the FEAST algorithm with inexact system solves.
- A Rational Function Preconditioner For Indefinite Sparse Linear Systems
- Computational materials science and engineering
- Coupled FE-BE method for eigenvalue analysis of elastic structures submerged in an infinite fluid domain
- ChASE: Chebyshev accelerated subspace iteration eigensolver for sequences of Hermitian eigenvalue problems
- Beyond automated multilevel substructuring: domain decomposition with rational filtering
- Dissecting the FEAST algorithm for generalized eigenproblems
Uses Software
This page was built for publication: FEAST as a subspace iteration eigensolver accelerated by approximate spectral projection
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2923348)
