Competition in high dimensional spaces using a sparse approximation of neural fields

Jean-Charles Quinton 1 Bernard Girau 1 Mathieu Lefort 1
1 CORTEX - Neuromimetic intelligence
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : The Continuum Neural Field Theory implements competition within topologically organized neural networks with lateral inhibitory connections. However, due to the polynomial complexity of matrix-based implementations, updating dense representations of the activity becomes computationally intractable when an adaptive resolution or an arbitrary number of input dimensions is required. This paper proposes an alternative to self-organizing maps with a sparse implementation based on Gaussian mixture models, promoting a trade-off in redundancy for higher computational efficiency and alleviating constraints on the underlying substrate. This version reproduces the emergent attentional properties of the original equations, by directly applying them within a continuous approximation of a high dimensional neural field. The model is compatible with preprocessed sensory flows but can also be interfaced with artificial systems. This is particularly important for sensorimotor systems, where decisions and motor actions must be taken and updated in real-time. Preliminary tests are performed on a reactive color tracking application, using spatially distributed color features.
Type de document :
Chapitre d'ouvrage
Carlos Hernández, Ricardo Sanz, Jaime Gómez-Ramirez, Leslie S. Smith, Amir Hussain, Antonio Chella, Igor Aleksander. From Brains to Systems: Brain-Inspired Cognitive Systems 2010, Springer, 2011, Advances in Experimental Medicine and Biology, 978-1461401636
Liste complète des métadonnées

Littérature citée [33 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00568924
Contributeur : Jean-Charles Quinton <>
Soumis le : jeudi 24 février 2011 - 15:35:09
Dernière modification le : jeudi 7 juin 2018 - 09:52:03
Document(s) archivé(s) le : mercredi 25 mai 2011 - 02:30:06

Fichier

Quinton_BICS_10_Chapter.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00568924, version 1

Collections

Citation

Jean-Charles Quinton, Bernard Girau, Mathieu Lefort. Competition in high dimensional spaces using a sparse approximation of neural fields. Carlos Hernández, Ricardo Sanz, Jaime Gómez-Ramirez, Leslie S. Smith, Amir Hussain, Antonio Chella, Igor Aleksander. From Brains to Systems: Brain-Inspired Cognitive Systems 2010, Springer, 2011, Advances in Experimental Medicine and Biology, 978-1461401636. 〈inria-00568924〉

Partager

Métriques

Consultations de la notice

484

Téléchargements de fichiers

161