Identification of binary images by a vector neural network with a measure of affinity between conditions neurons

V. M. Kryzhanovskiy

Identification of binary images by a vector neural network with a measure of affinity between conditions neurons

V. M. Kryzhanovskiy¹

research article Pages: 33-46

¹ Научно-исследовательский институт системных исследований РАН, Москва

Keywords: identification of binary images vector neural networks stability of the interference

Abstract

The new model of a vector neural network with the aprioristic information on distribution of noise, for increase of a noise stability of a network is described.

The measure of affinity between conditions нейрона is entered, allowing to consider the aprioristic information, and to increase by the order capacity of the memory created on the basis of this model.

How to cite

Крыжановский В.М. Идентификация бинарных образов векторной нейронной сетью с мерой близости между состояниями нейронов // Нейрокомпьютеры: разработка, применение. 2011. № 4. С. 33–46.

References

Kanter, I.,Potts-glass models of neural networks. Physical Review A. 1988. V. 37(7). P. 2739-2742.
Cook, J., The mean-field theory of a Q-state neural network model. Journal of Physics A. 1989. V. 22. P. 2000-2012.
Nadal, J., Rau, A.,Storage capacity of a Potts-perceptron // J. Phys. I: France 1. 1991. P. 1109-1121.
Bolle, D., Dupont P., andHuyghebaert, J., Thermodynamics properties of the q-state Potts-glass neural network // Phys. Rew. A. 1992. V. 45. P. 4194-4197.
Kryzhanovsky, B.V., Kryzhanovsky, V. M., Mikaelian, A. L., Fonarev, A., Parametric dynamic neural network recognition power // Optical Memory & Neural Network. 2001. V. 10. № 4. P. 211-218.
Alieva, D. I., Kryzhanovsky, B. V., Kryzhanovsky, V. M., Q-valued neural network as a system of fast identification and pattern recognition // Pattern Recognition and Image Analysis. 2005. V. 15. № 1. P. 30-33.
Крыжановский В. М., Симкина Д. И. Свойства клиппированной модели векторной ассоциативной памяти // Вестник компьютерных информационных технологий. 2007. № 11. С. 20-25.
Kryzhanovsky, V. M.,Modified q-state Potts model with binarized synaptic coefficients // Lecture Notes in Computer Science. 2008. V. 5164. Part II. P. 72-80.
Kryzhanovsky, B. V., Simkina, D. I., Kryzhanovsky, V. M., A Vector Model of Associative Memory with Clipped Synapses // Pattern Recognition and Image Analysis. 2009. V. 19. № 2. P. 289-295.
Kryzhanovsky, B., Kryzhanovskiy V., Litinskii, L., Machine Learning in Vector Models of Neural Networks // Advances in Machine Learning II. Dedicated to the memory of Professor Ryszard S. Michalski. Koronacki, J., Ras, Z.W., Wierzchon, S.T. (et al.) (Eds.), Series "Studies in Computational Intelligence". Springer. ISSN: 1860-949X. SCI 263. 2010. P. 427-443.
Крыжановский Б. В., Крыжановский В. М. Идентификатор бинарных образов на основе модели Поттса // Вестник информационных компьютерных технологий. 2009. № 8. С. 24-30.
Kryzhanovsky, B.V., Kryzhanovsky, V. M., Fonarev, A. B., Decorrelating Parametrical Neural Network // Proc. of International Joint Conference on Neural Networks (IJCNN-2005). 2005. P. 1023-1026.
Kryzhanovsky, V. M., Kryzhanovsky, B. V., Fonarev, A. B., Application of Potts-model perceptron for binary patterns identification // ICANN, Lecture Notes in Computer Science 5163. Part I. 2008. P. 553-561.
Hopfield J. J.,Neural Networks and physical systems with emergent collective computational abilities // Proc.Nat.Acad.Sci.USA. 1982. V. 79. P. 2554-2558.

[1] Kanter, I.,Potts-glass models of neural networks. Physical Review A. 1988. V. 37(7). P. 2739-2742.

[2] Cook, J., The mean-field theory of a Q-state neural network model. Journal of Physics A. 1989. V. 22. P. 2000-2012.

[3] Nadal, J., Rau, A.,Storage capacity of a Potts-perceptron // J. Phys. I: France 1. 1991. P. 1109-1121.

[4] Bolle, D., Dupont P., andHuyghebaert, J., Thermodynamics properties of the q-state Potts-glass neural network // Phys. Rew. A. 1992. V. 45. P. 4194-4197.

[5] Kryzhanovsky, B.V., Kryzhanovsky, V. M., Mikaelian, A. L., Fonarev, A., Parametric dynamic neural network recognition power // Optical Memory & Neural Network. 2001. V. 10. № 4. P. 211-218.

[6] Alieva, D. I., Kryzhanovsky, B. V., Kryzhanovsky, V. M., Q-valued neural network as a system of fast identification and pattern recognition // Pattern Recognition and Image Analysis. 2005. V. 15. № 1. P. 30-33.

[7] Крыжановский В. М., Симкина Д. И. Свойства клиппированной модели векторной ассоциативной памяти // Вестник компьютерных информационных технологий. 2007. № 11. С. 20-25.

[8] Kryzhanovsky, V. M.,Modified q-state Potts model with binarized synaptic coefficients // Lecture Notes in Computer Science. 2008. V. 5164. Part II. P. 72-80.

[9] Kryzhanovsky, B. V., Simkina, D. I., Kryzhanovsky, V. M., A Vector Model of Associative Memory with Clipped Synapses // Pattern Recognition and Image Analysis. 2009. V. 19. № 2. P. 289-295.

[10] Kryzhanovsky, B., Kryzhanovskiy V., Litinskii, L., Machine Learning in Vector Models of Neural Networks // Advances in Machine Learning II. Dedicated to the memory of Professor Ryszard S. Michalski. Koronacki, J., Ras, Z.W., Wierzchon, S.T. (et al.) (Eds.), Series "Studies in Computational Intelligence". Springer. ISSN: 1860-949X. SCI 263. 2010. P. 427-443.

[11] Крыжановский Б. В., Крыжановский В. М. Идентификатор бинарных образов на основе модели Поттса // Вестник информационных компьютерных технологий. 2009. № 8. С. 24-30.

[12] Kryzhanovsky, B.V., Kryzhanovsky, V. M., Fonarev, A. B., Decorrelating Parametrical Neural Network // Proc. of International Joint Conference on Neural Networks (IJCNN-2005). 2005. P. 1023-1026.

[13] Kryzhanovsky, V. M., Kryzhanovsky, B. V., Fonarev, A. B., Application of Potts-model perceptron for binary patterns identification // ICANN, Lecture Notes in Computer Science 5163. Part I. 2008. P. 553-561.

[14] Hopfield J. J.,Neural Networks and physical systems with emergent collective computational abilities // Proc.Nat.Acad.Sci.USA. 1982. V. 79. P. 2554-2558.