This work presents an evolutionary approach for the optimization of neural networks design, based on the joint evolution of the topology and the connection weights, providing a novel similarity-based crossover that aims to overcome one of the major problems of this operator, known as the permutation problem. The approach has been implemented and applied to two benchmark classification problems in machine learning, and the experimental results, compared to those obtained by other works in the literature, show how it can produce compact neural networks with a satisfactory generalization capability.
A novel similarity-based crossover for artificial neural network evolution / A. Azzini, M. Dragoni, A.G.B. Tettamanzi - In: Parallel problem solving from nature, PPSN XI : 11th international conference, Krakow, Poland, september 11-15, 2010 : proceedings. Part 1. / [a cura di] R. Schaefer ... [et al.]. - Berlin : Springer, 2010. - ISBN 9783642158438. - pp. 344-353 (( Intervento presentato al 11. convegno International Conference on Parallel Problem Solving from Nature (PPSN) tenutosi a Krakow, Poland nel 2010 [10.1007/978-3-642-15844-5_35].
A novel similarity-based crossover for artificial neural network evolution
A. AzziniPrimo
;M. DragoniSecondo
;A.G.B. TettamanziUltimo
2010
Abstract
This work presents an evolutionary approach for the optimization of neural networks design, based on the joint evolution of the topology and the connection weights, providing a novel similarity-based crossover that aims to overcome one of the major problems of this operator, known as the permutation problem. The approach has been implemented and applied to two benchmark classification problems in machine learning, and the experimental results, compared to those obtained by other works in the literature, show how it can produce compact neural networks with a satisfactory generalization capability.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.