Dictionary descent in optimization
Machine Learning
2015-11-05 v1 Numerical Analysis
Abstract
The problem of convex optimization is studied. Usually in convex optimization the minimization is over a d-dimensional domain. Very often the convergence rate of an optimization algorithm depends on the dimension d. The algorithms studied in this paper utilize dictionaries instead of a canonical basis used in the coordinate descent algorithms. We show how this approach allows us to reduce dimensionality of the problem. Also, we investigate which properties of a dictionary are beneficial for the convergence rate of typical greedy-type algorithms.
Cite
@article{arxiv.1511.01304,
title = {Dictionary descent in optimization},
author = {Vladimir Temlyakov},
journal= {arXiv preprint arXiv:1511.01304},
year = {2015}
}
Comments
arXiv admin note: text overlap with arXiv:1206.0392