Showing 1–1 of 1 results for author: Gutiérrez-Peña, P A

Distribution of the search of evolutionary product unit neural networks for classification

Authors: A. J. Tallón-Ballesteros, P. A. Gutiérrez-Peña, C. Hervás-Martínez

Abstract: This paper deals with the distributed processing in the search for an optimum classification model using evolutionary product unit neural networks. For this distributed search we used a cluster of computers. Our objective is to obtain a more efficient design than those net architectures which do not use a distributed process and which thus result in simpler designs. In order to get the best classi… ▽ More This paper deals with the distributed processing in the search for an optimum classification model using evolutionary product unit neural networks. For this distributed search we used a cluster of computers. Our objective is to obtain a more efficient design than those net architectures which do not use a distributed process and which thus result in simpler designs. In order to get the best classification models we use evolutionary algorithms to train and design neural networks, which require a very time consuming computation. The reasons behind the need for this distribution are various. It is complicated to train this type of nets because of the difficulty entailed in determining their architecture due to the complex error surface. On the other hand, the use of evolutionary algorithms involves running a great number of tests with different seeds and parameters, thus resulting in a high computational cost △ Less

Submitted 15 May, 2012; originally announced May 2012.

Comments: 8 pages, 2 figures, in Proc. IADIS International Conference Applied Computing 2007 (AC 2007), ISBN 978-972-8924-30-0, pp. 266-273, Spain. Note: "This is a reprint from a paper published in the Proceedings of the IADIS International Conference Applied Comupting 2007, http://www.iadis.org"