Determination of hyper-parameters for kernel based classification and regression
Loading...
Date
2005-11-07T11:39:49Z
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
We investigate methods to determine appropriate choices of the hyper-parameters for kernel based methods. Support vector classification, kernel logistic regression and support vector regression are considered. Grid search, Nelder-Mead algorithm
and pattern search algorithm are used.
Description
Table of contents
Keywords
convex risk minimization, kernel logistic regression, statistical machine learning, support vector machine, support vector regression