LSRN: A parallel iterative solver for strongly over- or underdetermined systems (Q2875015)
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: LSRN: A parallel iterative solver for strongly over- or underdetermined systems |
scientific article; zbMATH DE number 6329840
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | LSRN: A parallel iterative solver for strongly over- or underdetermined systems |
scientific article; zbMATH DE number 6329840 |
Statements
13 August 2014
0 references
linear least squares
0 references
overdetermined system, underdetermined system, rank-deficient
0 references
minimum-length solution
0 references
sparse matrix
0 references
iterative method
0 references
preconditioning
0 references
Chebyshev semi-iterative method
0 references
Tikhonov regularization
0 references
ridge regression
0 references
parallel computing
0 references
random projection
0 references
random sampling
0 references
random matrix
0 references
randomized algorithm
0 references
LSRN: A parallel iterative solver for strongly over- or underdetermined systems (English)
0 references
The authors propose a random normal projection least squares solver for general linear least squares problems and their Tikhonov regularized versions. The method iteratively constructs a provably efficient preconditioner. It has a fully predictable run-time performance and scales well in parallel environments.
0 references
0.7910975217819214
0 references
0.7775701880455017
0 references
0.772622287273407
0 references
0.7716783881187439
0 references
0.7704228758811951
0 references
