Catalogo Articoli (Spogli Riviste)
OPAC HELP
Titolo: Layered neural network with intralayer connections using Qstates clock neurons
Autore: Katayama, K; Horiguchi, T;
 Indirizzi:
 Tohoku Univ, GSIS, Dept Comp & Math Sci, Sendai, Miyagi 9808579, Japan Tohoku Univ Sendai Miyagi Japan 9808579 ci, Sendai, Miyagi 9808579, Japan
 Titolo Testata:
 PHYSICA A
fascicolo: 34,
volume: 297,
anno: 2001,
pagine: 532  548
 SICI:
 03784371(20010815)297:34<532:LNNWIC>2.0.ZU;23
 Fonte:
 ISI
 Lingua:
 ENG
 Soggetto:
 RANDOM SYNAPTIC DILUTION; SPINGLASS; STATISTICALMECHANICS; NATURAL FREQUENCIES; ASSOCIATIVE MEMORY; DYNAMICS; MODEL; PATTERNS; SATURATION; CORTEX;
 Keywords:
 neural network; Hebb rule; Qstates clock neuron; oscillatory neuron; storage capacity; replica method;
 Tipo documento:
 Article
 Natura:
 Periodico
 Settore Disciplinare:
 Physical, Chemical & Earth Sciences
 Citazioni:
 29
 Recensione:
 Indirizzi per estratti:
 Indirizzo: Horiguchi, T Tohoku Univ, GSIS, Dept Comp & Math Sci, Sendai, Miyagi 9808579, Japan Tohoku Univ Sendai Miyagi Japan 9808579 iyagi 9808579, Japan



 Citazione:
 K. Katayama e T. Horiguchi, "Layered neural network with intralayer connections using Qstates clock neurons", PHYSICA A, 297(34), 2001, pp. 532548
Abstract
We investigate storage capacity of a fully connected layered neural network with Q(greater than or equal to 2)states clock neurons, including Q = infinity (corresponding to oscillatory neurons) and with intralayer connections, where random Qvalues patterns are embedded into the network by the Hebbian learning rule. We assume that both of interlayered neurons and intralayered neurons are updated simultaneously. We are able to treat a recurrent network, a feedforward network and a layered network with the intralayer connections within the same framework by adjusting a parameter. We analyze an energy function of the network by using the replica method. We clarify that the storage capacity of the layered network with the intralaver connections is enhanced in comparison with that of the recurrent network and that of the feedforward network for Q = 2,3,4 and 5 at zero temperature; this is due to the competition between the interlayer connections and the intralayer connections. For Q greater than or equal to 6, including Q = infinity, the storage capacity at zero temperature is maximum for the feedforward network. We obtain phase diagrams in a loadparameter versus temperature plane for the layered network with the intralayer connections. (C) 2001 Elsevier Science B.V. All rights reserved.
ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 15/07/20 alle ore 14:46:43