Evolving Sum and Composite Kernel Functions for Regularization Networks

Authors: Petra Vidnerová, Roman Neruda
Publication year: 2011
Publication type: Konferenční příspěvek (mezinárodní konf.)
Periodical: Adaptive and Natural Computing Algorithms. Part I
Projects: Analysis of Intelligent Computational Distributed Systems
Neural Networks Learning Algorithms Based on Regularization Theory
Bibliographic citation: VIDNEROVÁ, Petra, NERUDA, Roman. Evolving Sum and Composite Kernel Functions for Regularization Networks. In: DOBNIKAR, A., LOTRIČ, U., ŠTER, B., eds. Adaptive and Natural Computing Algorithms. Part I. Vol. 1. Heidelberg: Springer, 2011, S. 180-189. Lecture Notes in Computer Science, 6593. ISBN 978-3-642-20281-0. ISSN 0302-9743
Dostupné z: doi:10.1007/978-3-642-20282-7_19
DOI: 10.1007/978-3-642-20282-7_19
ID: 359155

Abstract

In this paper we propose a novel evolutionary algorithm for regularization networks. The main drawback of regularization networks in practical applications is the presence of meta-parameters, including the type and parameters of kernel functions Our learning algorithm provides a solution to this problem by searching through a space of different kernel functions, including sum and composite kernels. Thus, an optimal combination of kernel functions with parameters is evolved for given task specified by training data. Comparisons of composite kernels, single kernels, and traditional Gaussians are provided in several experiments.