Hybrid Learning of Regularization Neural Networks

0345012 - ÚI 2011 RIV DE eng C - Conference Paper (international conference)
Vidnerová, Petra - Neruda, Roman
Hybrid Learning of Regularization Neural Networks.
Artificial Intelligence and Soft Computing. Vol. 2. Berlin: Springer, 2010 - (Rutkowski, L.; Scherer, R.; Tadeusiewicz, R.; Zadeh, L.; Zurada, J.), s. 124-131. Lecture Notes in Artificial Intelligence, 6114. ISBN 978-3-642-13231-5. ISSN 0302-9743.
[ICAISC 2010. International Conference on Artifical Intelligence and Soft Computing /10./. Zakopane (PL), 13.06.2010-17.06.2010]
R&D Projects: GA AV ČR KJB100300804
Institutional research plan: CEZ:AV0Z10300504
Keywords : supervised learning * regularization networks * genetic algorithms
Subject RIV: IN - Informatics, Computer Science

Regularization theory presents a sound framework to solving supervised learning problems. However, the regularization networks have a large size corresponding to the size of training data. In this work we study a relationship between network complexity, i.e. number of hidden units, and approximation and generalization ability. We propose an incremental hybrid learning algorithm that produces smaller networks with performance similar to original regularization networks.
Permanent Link: http://hdl.handle.net/11104/0186392