Low-rank matrix completion by Riemannian optimization (Q2848192)
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: Low-rank matrix completion by Riemannian optimization |
scientific article; zbMATH DE number 6211579
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Low-rank matrix completion by Riemannian optimization |
scientific article; zbMATH DE number 6211579 |
Statements
25 September 2013
0 references
matrix completion
0 references
low-rank matrices
0 references
optimization on manifolds
0 references
differential geometry
0 references
nonlinear conjugate gradients
0 references
Riemannian manifolds
0 references
Low-rank matrix completion by Riemannian optimization (English)
0 references
The matrix completion problem consists basically of recovering a low-rank matrix based on a very sparse set of entries of this matrix. In this paper, the author proposes a method based on optimization on manifolds to solve large-scale matrix completion problems. The approach consists of directly minimizing the least-squares error of the fit directly over the set of matrices of a given rank. The method proposed presents however several pertinent limitations.
0 references
0.892124593257904
0 references
0.8767210841178894
0 references
0.8736577033996582
0 references
0.8511332869529724
0 references
0.8510798811912537
0 references
