Expressive Power of Non-deterministic Evolving Recurrent Neural Networks in Terms of Their Attractor Dynamics

Détails

ID Serval
serval:BIB_3FC9ED926E81
Type
Article: article d'un périodique ou d'un magazine.
Collection
Publications
Institution
Titre
Expressive Power of Non-deterministic Evolving Recurrent Neural Networks in Terms of Their Attractor Dynamics
Périodique
Unconventional Computation and Natural Computation
Auteur(s)
Cabessa J., Duparc J.
ISBN
9783319218182
9783319218199
ISSN
0302-9743
1611-3349
Statut éditorial
Publié
Date de publication
2015
Peer-reviewed
Oui
Volume
12
Numéro
1
Pages
144-156
Langue
anglais
Résumé
We provide a characterization of the expressive powers of several models of nondeterministic recurrent neural networks according to their attractor dynamics. More precisely, we consider two forms of nondeterministic neural networks. In the first case, nondeterminism is expressed as an external binary guess stream processed by means of an additional Boolean guess cell. In the second case, nondeterminism is expressed as a set of possible evolving patterns that the synaptic connections of the network might follow over the successive time steps. In these two contexts, ten models of nondeterministic neural networks are considered, according to the nature of their synaptic weights. Overall, we prove that the static rational-weighted neural networks of type 1 are computationally equivalent to nondeterministic Muller Turing machines. They recognize the class of all effectively analytic (Sigma(1)(1) lightface) sets. The nine other models of analog and/or evolving neural networks of types 1 and 2 are all computationally equivalent to each other, and strictly more powerful than nondeterministic Muller Turing machines. They recognize the class of all analytic (Sigma(1)(1) boldface) sets.
Mots-clé
Recurrent neural networks, Neural computation, Analog computation, Evolving systems, Attractors, Spatiotemporal patterns, Turing machines, Expressive power, Omega-languages, Borel sets, Analytic sets
Web of science
Création de la notice
10/05/2017 14:29
Dernière modification de la notice
20/08/2019 14:37
Données d'usage