No label defined (Q4027344)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 125628
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 125628 |
Statements
21 February 1993
0 references
complexity
0 references
orthogonalization
0 references
iterative methods
0 references
large symmetric systems
0 references
eigenvalue problems
0 references
matrix factorization
0 references
Gram-Schmidt method
0 references
0.796720027923584
0 references
0.7839692234992981
0 references
0.7838993072509766
0 references
0.7829903960227966
0 references
[For the entire collection see Zbl 0745.00024.] The orthogonalization of a set of vectors, with respect to an \(M\)-norm given by the inner product \(u^ TMu\) (\(M=B^ TB\) positive definite matrix), is a basic step in iterative methods for solving large symmetric systems of linear equations and for generalized eigenvalue problems. It is shown that the matrix factorization \(S=GR\), where the \((p,n)\)- matrix \(G\) is \(M\)-orthogonal and \(R\) is upper triangular, can be obtained efficiently by a generalized Gram-Schmidt method. An a posteriori method of measuring the loss of orthogonality in the columns of the computed matrix \(G\) is discussed.
0 references
