RefDoc
Haut

Faire une nouvelle recherche
Make a new search
Lancer la recherche


Titre du document / Document title

The Kernel Hopfield memory network

Auteur(s) / Author(s)

GARCIA Cristina (1) ; MORENO José Ali (1) ;

Affiliation(s) du ou des auteurs / Author(s) Affiliation(s)

(1) Laboratorio de Computación Emergente, Facultades de Ciencias e Ingeniería, Universidad Central de Venezuela, VENEZUELA

Résumé / Abstract

The kernel theory drawn from the work on learning machines is applied to the Hopfield neural network. This provides a new insight into the workings of the neural network as associative memory. The kernel trick defines an embedding of memory patterns into (higher or infinite dimensional) memory feature vectors and the training of the network is carried out in this feature space. The generalization of the network by using the kernel theory improves its performance in three aspects. First, an adequate kernel selection enables the satisfaction of the condition that any set of memory patterns be attractors of the network dynamics. Second, the basins of attraction of the memory patterns are enhanced improving the recall capacity. Third, since the memory patterns are mapped into a higher dimensional feature space the memory capacity density is effectively increased. These aspects are experimentally demonstrated on sets of random memory patterns.

Revue / Journal Title

Lecture notes in computer science    ISSN  0302-9743 

Source / Source

Congrès
Cellular automata :   ( Amsterdam, 25-27 October 2004 )
ACRI 2004 : international conference on cellular automata for research and industry No6, Amsterdam , PAYS-BAS (25/10/2004)
2004  , vol. 3305, pp. 755-764[Note(s) : XV, 883 p., ] [Document : 10 p.] (12 ref.) ISBN 3-540-23596-5 ;  Illustration : Illustration ;

Langue / Language

Anglais

Editeur / Publisher

Springer, Berlin, ALLEMAGNE  (1973) (Revue)
Springer, Berlin, ALLEMAGNE  (2004) (Monographie)

Mots-clés anglais / English Keywords

Random set

;

Memory capacity

;

Attraction

;

Attractor

;

Neural network

;

Hopfield model

;

Hopfield neural nets

;

Associative memory

;

Artificial intelligence

;

Cellular automaton

;

Mots-clés français / French Keywords

Ensemble aléatoire

;

Capacité mémoire

;

Attraction

;

Attracteur

;

Réseau neuronal

;

Modèle Hopfield

;

Réseau neuronal Hopfield

;

Mémoire associative

;

Intelligence artificielle

;

Automate cellulaire

;

Mots-clés espagnols / Spanish Keywords

Conjunto aleatorio

;

Capacidad memoria

;

Atracción

;

Atractor

;

Red neuronal

;

Modelo Hopfield

;

Memoria asociativa

;

Inteligencia artificial

;

Autómata celular

;

Localisation / Location

INIST-CNRS, Cote INIST : 16343, 35400012437928.0780

Nº notice refdoc (ud4) : 16334395



Faire une nouvelle recherche
Make a new search
Lancer la recherche
Bas