Coordinate and subspace optimization methods for linear least squares with non-quadratic regularization (Q2465768): Difference between revisions
From MaRDI portal
Changed an Item |
Changed an Item |
||
Property / describes a project that uses | |||
Property / describes a project that uses: PDCO / rank | |||
Normal rank |
Revision as of 01:13, 28 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Coordinate and subspace optimization methods for linear least squares with non-quadratic regularization |
scientific article |
Statements
Coordinate and subspace optimization methods for linear least squares with non-quadratic regularization (English)
0 references
8 January 2008
0 references
The authors consider the problem of linear least squares with non-quadratic regularization (RLS) in the form: minimize \(\{ \|Az - b\|^2 + \rho (z) \}\), for \(z \in\mathbb R^n\), where \(A\) is an \(m \times n, (m < n)\) full rank matrix. They prove that for this problem the parallel coordinate descent algorithm (previously proposed by one of the authors), the expectation-maximization one and a shrinkage minimization method are essentially equivalent for the numerical solution of the RLS problem. Some image denoising numerical experiments are also presented.
0 references
least squares
0 references
regularization
0 references
inverse problems
0 references
denoising
0 references
shrinkage
0 references
basis pursuit
0 references
sparsity
0 references
coordinate-descent
0 references
proximal point
0 references
numerical experiments
0 references