The simulation of spin glass model of neural networks by the Wang-Landau algorithm

F. Yaşar; M. Dilaver

doi:10.1016/j.cpc.2007.05.017

The simulation of spin glass model of neural networks by the Wang-Landau algorithm

Araştırma sonucu: Dergiye katkı › Makale › bilirkişi

3 Alıntılar (Scopus)

Özet

In the present work, the Hopfield neural network model with infinite-range interactions was simulated by using the Wang-Landau algorithm. All simulations and measurements were done in the replica symmetric and broken regions of the model with discrete ∓1 values of random variables. The physical quantities such as energy density, specific heat were evaluated at all temperatures. Our results are in good agreement with the replica symmetric mean field solutions.

Orijinal dil	İngilizce
Sayfa (başlangıç-bitiş)	625-630
Sayfa sayısı	6
Dergi	Computer Physics Communications
Hacim	177
Basın numarası	8
DOI'lar	https://doi.org/10.1016/j.cpc.2007.05.017
Yayın durumu	Yayınlandı - 15 Eki 2007

Belgeye Erişim

10.1016/j.cpc.2007.05.017

Diger dosyalar ve linkler

Link to publication in Scopus

Bundan alıntı yap

@article{6da6a4fab88a44c2a9f608c5342c63f5,

title = "The simulation of spin glass model of neural networks by the Wang-Landau algorithm",

abstract = "In the present work, the Hopfield neural network model with infinite-range interactions was simulated by using the Wang-Landau algorithm. All simulations and measurements were done in the replica symmetric and broken regions of the model with discrete ∓1 values of random variables. The physical quantities such as energy density, specific heat were evaluated at all temperatures. Our results are in good agreement with the replica symmetric mean field solutions.",

keywords = "Hopfield model, Neural network, Spin glasses, Wang-Landau algorithm",

author = "F. Ya{\c s}ar and M. Dilaver",

year = "2007",

month = oct,

day = "15",

doi = "10.1016/j.cpc.2007.05.017",

language = "English",

volume = "177",

pages = "625--630",

journal = "Computer Physics Communications",

issn = "0010-4655",

publisher = "Elsevier B.V.",

number = "8",

}

TY - JOUR

T1 - The simulation of spin glass model of neural networks by the Wang-Landau algorithm

AU - Yaşar, F.

AU - Dilaver, M.

PY - 2007/10/15

Y1 - 2007/10/15

N2 - In the present work, the Hopfield neural network model with infinite-range interactions was simulated by using the Wang-Landau algorithm. All simulations and measurements were done in the replica symmetric and broken regions of the model with discrete ∓1 values of random variables. The physical quantities such as energy density, specific heat were evaluated at all temperatures. Our results are in good agreement with the replica symmetric mean field solutions.

AB - In the present work, the Hopfield neural network model with infinite-range interactions was simulated by using the Wang-Landau algorithm. All simulations and measurements were done in the replica symmetric and broken regions of the model with discrete ∓1 values of random variables. The physical quantities such as energy density, specific heat were evaluated at all temperatures. Our results are in good agreement with the replica symmetric mean field solutions.

KW - Hopfield model

KW - Neural network

KW - Spin glasses

KW - Wang-Landau algorithm

UR - https://www.scopus.com/pages/publications/34548671866

U2 - 10.1016/j.cpc.2007.05.017

DO - 10.1016/j.cpc.2007.05.017

M3 - Article

AN - SCOPUS:34548671866

SN - 0010-4655

VL - 177

SP - 625

EP - 630

JO - Computer Physics Communications

JF - Computer Physics Communications

IS - 8

ER -

The simulation of spin glass model of neural networks by the Wang-Landau algorithm

Özet

Belgeye Erişim

Diger dosyalar ve linkler

Parmak izi

Bundan alıntı yap