A coordinate gradient descent method for nonsmooth separable minimization (Q959979)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A coordinate gradient descent method for nonsmooth separable minimization |
scientific article |
Statements
A coordinate gradient descent method for nonsmooth separable minimization (English)
0 references
16 December 2008
0 references
The authors presented a block coordinate gradient descent method for minimizing the sum of a smooth function and a convex separable function. The method may be viewed as a hybrid of gradient-projection and coordinate descent methods, or as a block coordinate version of descent methods in [\textit{J. V. Burke}, Math. Program. 33, 260--279 (1985; Zbl 0581.90084)] and [\textit{M. Fukushima} and \textit{H. Mine}, Int. J. Syst. Sci. 12, 989--1000 (1981; Zbl 0467.65028)]. The global convergence is proved and, under a local Lipschitzian error bound assumption, linear convergence of this method. The local Lipschitzian error bound holds under assumptions analogous to those for constrained smooth optimization, e.g. the convex function is polyhedral and the smooth function is (nonconvex) quadratic or is the composition of a strongly convex function with a linear mapping. The numerical results are presented to verify the practical efficiency of the method.
0 references
error bound
0 references
global convergence
0 references
linear convergence rate
0 references
nonsmooth optimization
0 references
coordinate descent
0 references