Randomized Krylov Subspace Methods for Shifted Linear Systems

DOI10.48550/ARXIV.2610.34964MaRDI QIDQ7408822arXivFDO

Authors Xiang Yuan, Gang Wu

Publication date 26 October 2026

Published in arXiv

Tldr We develop randomized Krylov subspace methods for shifted linear systems using a unified framework that combines multiple advanced techniques, including randomized sketching, control of collinear residual directions, and deflated restarting. These novel algorithms reduce computational overhead while preserving orthogonality and achieve moderate improvements in convergence speed compared to traditional algorithms.

Full work available at URL https://doi.org/10.48550/ARXIV.2610.34964

arXiv classification

math.NA

Cites work

• A harmonic restarted Arnoldi algorithm for calculating eigenvalues and determining multiplicity
• A new shifted block GMRES method with inexact breakdowns for solving multi-shifted and multiple right-hand sides linear systems
• A Parallel and Vector Variant of the Cyclic Reduction Algorithm
• A preconditioned and shifted GMRES algorithm for the PageRank problem with multiple damping factors
• A shifted block FOM algorithm with deflated restarting for matrix exponential computations
• A sketch-and-select Arnoldi process
• An Iterative Method for Nonsymmetric Systems with Multiple Right-Hand Sides
• Arnoldi methods for large Sylvester-like observer matrix equations, and an associated algorithm for partial spectrum assignment
• Computing the action of the matrix exponential, with an application to exponential integrators
• Efficient and stable Arnoldi restarts for matrix functions based on quadrature
• Fast and accurate randomized algorithms for linear systems and eigenvalue problems
• GMRES with Deflated Restarting
• GMRES with randomized sketching and deflated restarting
• GMRES: A Generalized Minimal Residual Algorithm for Solving Nonsymmetric Linear Systems
• Implicitly Restarted GMRES and Arnoldi Methods for Nonsymmetric Systems of Equations
• Iterative Methods for Sparse Linear Systems
• Matrix methods in data mining and pattern recognition
• Numerical linear algebra aspects of control design computations
• On the use of matrix functions for fractional partial differential equations
• Preserving geometric properties of the exponential matrix by block Krylov subspace methods
• Randomized block Gram-Schmidt process for the solution of linear systems and eigenvalue problems
• Randomized flexible GMRES with deflated restarting
• Randomized Gram-Schmidt process with application to GMRES
• Randomized Numerical Linear Algebra: Foundations & Algorithms
• Randomized Orthogonal Projection Methods for Krylov Subspace Solvers
• Restarted full orthogonalization method for shifted linear systems
• Restarted GMRES for Shifted Linear Systems
• Sketching as a tool for numerical linear algebra
• Talbot quadratures and rational approximations
• The University of Florida sparse matrix collection
• Weighted Inner Products for GMRES and GMRES-DR

This page was built for publication: Randomized Krylov Subspace Methods for Shifted Linear Systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7408822)