Evolutionary Feature Selection for Spiking Neural Network Pattern Classifiers

Michal Valko 1 Nuno Cavalheiro 2 Marco Castelani 2
1 Department of Artificial Intelligence, Institute of Applied Informatics
2 FCT-UNL - Faculdade de Ciências e Technologia [Lisboa]
Abstract : This paper presents an application of the biologically realistic JASTAP neural network model to classification tasks. The JASTAP neural network model is presented as an alternative to the basic multi-layer perceptron model. An evolutionary procedure previously applied to the simultaneous solution of feature selection and neural network training on standard multi-layer perceptrons is extended with JASTAP model. Preliminary results on IRIS standard data set give evidence that this extension allows the use of smaller neural networks that can handle noisier data without any degradation in classification accuracy.
Type de document :
Communication dans un congrès
Bento et al. Proceedings of 2005 Portuguese Conference on Artificial Intelligence, Dec 2005, Covilha, Portugal. IEEE, pp.181-187, 2005, 〈10.1109/EPIA.2005.341291〉
Liste complète des métadonnées

Littérature citée [20 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/hal-00643498
Contributeur : Michal Valko <>
Soumis le : mardi 22 novembre 2011 - 10:09:43
Dernière modification le : lundi 18 mars 2019 - 18:36:51
Document(s) archivé(s) le : vendredi 16 novembre 2012 - 11:40:38

Fichier

VMC05.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Michal Valko, Nuno Cavalheiro, Marco Castelani. Evolutionary Feature Selection for Spiking Neural Network Pattern Classifiers. Bento et al. Proceedings of 2005 Portuguese Conference on Artificial Intelligence, Dec 2005, Covilha, Portugal. IEEE, pp.181-187, 2005, 〈10.1109/EPIA.2005.341291〉. 〈hal-00643498〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

120

Téléchargements de fichiers

217

Contact


archive-ouverte@inria.fr