Efficient and stable Arnoldi restarts for matrix functions based on quadrature (Q2923364)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Efficient and stable Arnoldi restarts for matrix functions based on quadrature |
scientific article; zbMATH DE number 6356191
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Efficient and stable Arnoldi restarts for matrix functions based on quadrature |
scientific article; zbMATH DE number 6356191 |
Statements
15 October 2014
0 references
matrix function
0 references
integral expression of the error
0 references
Gaussian quadrature
0 references
restart
0 references
Krylov subspace approximation
0 references
restarted Arnoldi method
0 references
restarted Lanczos method
0 references
deflated restarting
0 references
polynomial interpolation
0 references
Padé approximation
0 references
numerical test
0 references
0.9038173
0 references
0.8823811
0 references
0 references
0.8759383
0 references
0.8750809
0 references
0 references
0.8706089
0 references
0.8654986
0 references
0.86534363
0 references
Efficient and stable Arnoldi restarts for matrix functions based on quadrature (English)
0 references
This paper deals with the computation of \(f(A)b\), the action of a matrix function on a vector. An integral representation is presented for the error of the iterates in the Arnoldi method. Then an quadrature-based restarting Arnoldi algorithm is derived. This new method is applicable for a large class of functions, requires no a priori spectral information, and runs with essentially constant computational work per restart cycle. Numerical tests are presented to show the efficiency and numerical stability of the proposed method.
0 references
