Improved matrix algorithms via the subsampled randomized Hadamard transform (Q2866237)
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: Improved matrix algorithms via the subsampled randomized Hadamard transform |
scientific article; zbMATH DE number 6238087
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Improved matrix algorithms via the subsampled randomized Hadamard transform |
scientific article; zbMATH DE number 6238087 |
Statements
13 December 2013
0 references
low-rank approximation
0 references
randomized algorithm
0 references
Hadamard transform
0 references
least squares regression
0 references
0.9294478
0 references
0.86912763
0 references
0.8656285
0 references
0.8585838
0 references
0.8579979
0 references
0.8569864
0 references
0 references
0.84857464
0 references
0.84789443
0 references
0.84685445
0 references
Improved matrix algorithms via the subsampled randomized Hadamard transform (English)
0 references
Several recent randomized linear algebra algorithms rely upon fast dimension reduction methods. A popular choice is the subsampled randomized Hadamard transform (SRHT). The authors address the efficacy of an SRHT-based low-rank matrix approximation technique introduced by \textit{F. Woolfe} et al. [Appl. Comput. Harmon. Anal. 25, No. 3, 335--366 (2008; Zbl 1155.65035)]. Some sharper approximation bounds are proved. They also discuss two approaches to least squares regression involving SRHT dimensionality reduction.
0 references
