Normalized iterative hard thresholding for matrix completion (Q2870672)
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 6248317
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Normalized iterative hard thresholding for matrix completion |
scientific article; zbMATH DE number 6248317 |
Statements
21 January 2014
0 references
matrix completion
0 references
compressed sensing
0 references
low rank approximation
0 references
alternating projection
0 references
algorithm
0 references
Normalized iterative hard thresholding for matrix completion (English)
0 references
Matrices of low rank can be uniquely determined from fewer linear measurements, or entries, than the total number of entries in the matrix. The authors propose an alternating projection algorithm which uses an adaptive step size calculated to be exact for a restricted subspace. The proposed method has near-optimal order recovery guarantees from dense measurement masks and average case performance superior in some respects to other matrix completion algorithms for dense measurement masks and entry measurements. The proposed algorithm is able to recover matrices from extremely close to the minimum number of measurements necessary.
0 references
0.8582205176353455
0 references
0.845502495765686
0 references
0.843479335308075
0 references
0.8381890654563904
0 references
