Geodesic
Method of pattern recognition based on the Boltzmann machine model
News of the Kabardin-Balkar scientific center of RAS, no. 6-2 (2015), pp. 54-60.

Voir la notice de l'article provenant de la source Math-Net.Ru

This paper proposes an approach to solving the problem of pattern recognition based on neural networks using techniques of "deep learning". Various types of models, such as restricted Boltzmann machine, deep Boltzmann machine are studied. An algorithm for training the models in question is described. Computational experiments were conducted to determine the effectiveness of the approach developed. Advantages and disadvantages of the proposed method are identified.
Keywords: pattern recognition, artificial neural networks, Boltzmann machine, machine learning. 
@article{IZKAB_2015_6-2_a7,
     author = {L. A. Gladkov and N. V. Gladkova and A. N. Babynin and A. M. Ksalov},
     title = {Method of pattern recognition based on the {Boltzmann} machine model},
     journal = {News of the Kabardin-Balkar scientific center of RAS},
     pages = {54--60},
     publisher = {mathdoc},
     number = {6-2},
     year = {2015},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/IZKAB_2015_6-2_a7/}
}
TY  - JOUR
AU  - L. A. Gladkov
AU  - N. V. Gladkova
AU  - A. N. Babynin
AU  - A. M. Ksalov
TI  - Method of pattern recognition based on the Boltzmann machine model
JO  - News of the Kabardin-Balkar scientific center of RAS
PY  - 2015
SP  - 54
EP  - 60
IS  - 6-2
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/IZKAB_2015_6-2_a7/
LA  - ru
ID  - IZKAB_2015_6-2_a7
ER  - 
%0 Journal Article
%A L. A. Gladkov
%A N. V. Gladkova
%A A. N. Babynin
%A A. M. Ksalov
%T Method of pattern recognition based on the Boltzmann machine model
%J News of the Kabardin-Balkar scientific center of RAS
%D 2015
%P 54-60
%N 6-2
%I mathdoc
%U http://geodesic.mathdoc.fr/item/IZKAB_2015_6-2_a7/
%G ru
%F IZKAB_2015_6-2_a7
L. A. Gladkov; N. V. Gladkova; A. N. Babynin; A. M. Ksalov. Method of pattern recognition based on the Boltzmann machine model. News of the Kabardin-Balkar scientific center of RAS, no. 6-2 (2015), pp. 54-60. http://geodesic.mathdoc.fr/item/IZKAB_2015_6-2_a7/

[1] D. H. Ackley, G. E. Hinton, T. J. Sejnowski, “A Learning Algorithm for Boltzmann Machines”, Cognitive Science, 9 (1985), 147–169

[2] A. V. Nazarov, A. I. Loskutov, Neirosetevye algoritmy prognozirovaniya i optimizatsii sistem, Nauka i tekhnika, SPb, 2003

[3] G. E. Hinton, “Training products of experts by minimizing contrastive divergence”, Neural Computation NECO, 14:8 (2002), 1771–1800

[4] R. Salakhutdinov, G. E. Hinton, “Deep Boltzmann machines”, In Proceedings of the International Conference on Artificial Intelligence and Statistics, v. 5, 2009, 448–455 pp.

[5] G. E. Hinton, R. R. Salakhutdinov, “Reducing the dimensionality of data with neural networks”, In Science, 313(5786) (2006), 504–507

[6] V. V. Bova, L. A. Gladkov, A. A. Lezhebokov, “Problem-oriented algorithms of solutions search based on the methods of swarm intelligence”, World Applied Sciences Journal, 2013, no. 27 (9), 1201–1205

[7] L. A. Gladkov, V. V. Kureichik, Yu. A. Kravchenko, “Evolutionary Algorithm for Extremal Subsets Comprehension in Graphs”, World Applied Sciences Journal, 2013, no. 27 (9), 1212–1217