On GMRES-equivalent bounded operators (Q2706250)
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: On GMRES-equivalent bounded operators |
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | On GMRES-equivalent bounded operators |
scientific article |
Statements
19 March 2001
0 references
GMRES
0 references
Arnoldi's method
0 references
bounded operators
0 references
Hilbert space
0 references
iterative methods
0 references
Krylov subspace method
0 references
generalized minimal residual method
0 references
On GMRES-equivalent bounded operators (English)
0 references
The author studies the generalized minimal residual (GMRES) method applied to some operator equation \(Ax= r\) in a Hilbert space \(H\), where the operator \(A\in L(H)\) is supposed to be linear and bounded. At the \(k\)th step, the GMRES produces an approximate solution which minimizes the residual norm \(\|Ax-r\|\) over the Krylov subspace \(K^k(A,r):= \text{span}\{A^0 r,A^1r,\dots, A^{k-1}r\}\).NEWLINENEWLINENEWLINEThe author constructs a unitary operator \(U\), a nonnegative definite, bounded selfadjoint operator \(P\) and an indefinite, bounded selfadjoint operator \(T\) such that all corresponding Krylov subspace \(K^k(A, R)\), \(K^k(U,r)\), \(K^k(P, r)\) and \(K^k(T, r)\) of the same dimension are equal. It is clear that this result can be heavily used in the analysis of the convergence properties of the GMRES.NEWLINENEWLINENEWLINEThe author reformulates all these results for the matrix case that is of most practical interest from the point of view of solving systems of linear algebraic equations.
0 references