Evolution of Product Kernels for Regularization Networks

0369187 - ÚI 2013 DE eng C - Konferenční příspěvek (zahraniční konf.)
Vidnerová, Petra - Neruda, Roman
Evolution of Product Kernels for Regularization Networks.
Advanced Intelligent Computing. Berlin: Springer, 2011 - (Huang, D.; Gan, Y.; Bevilacqua, V.; Figueroa, J.), nevyšlo tiskem. Lecture Notes in Computer Science, 6838. ISBN 978-3-642-24727-9. ISSN 0302-9743.
[ICIC 2011. International Conference on Intelligent Computing. Zhengzhou (CN), 11.08.2011-14.08.2011]
Klíčová slova: genetic algorithms * kernel functions * regularization networks
Kód oboru RIV: IN - Informatika

Approximation problems formulated as regularized minimization problems with kernel-based stabilizers lead to solutions of the shape of linear combination of kernel functions. These can be expressed as one-hidden layer feed-forward neural network schemes, however, the rich possibilities of theoretical approach are usually not exploited in suitable learning algorithms. In this paper we focus on regularization networks with product kernels and propose an evolutionary learning algorithm utilizing genetic search for suitable parameters. The approach is experimentally tested on experiments.
Trvalý link: http://hdl.handle.net/11104/0203312