Modified q-State Potts Model with Binarized Synaptic Coefficients
Practical applications of q-state Potts models are complicated, as they require very large RAM (32N 2 q 2 bits, where N is the number of neurons and q is the number of the states of a neuron). In this work we examine a modified Potts model with binarized synaptic coefficients. The procedure of binarization allows one to make the required RAM 32 times smaller (N 2 q 2 bits), and the algorithm more than q times faster. One would expect that the binarization worsens the recognizing properties. However our analysis shows an unexpected result: the binarization procedure leads to the increase of the storage capacity by a factor of 2. The obtained results are in a good agreement with the results of computer simulations.
KeywordsRecognition Potts model storage capacity
Unable to display preview. Download preview PDF.
- 7.Kryzhanovsky, B.V., Mikaelyan, A.L.: On the Recognizing Ability of a Neural Network on Neurons with Parametric Transformation of Frequencies. Doklady Mathematics 65(2), 286–288 (2002)Google Scholar
- 9.Kryzhanovsky, B.V., Mikaelyan, A.L.: An associative memory capable of recognizing strongly correlated patterns. Doklady Mathematics 67(3), 455–459 (2003)Google Scholar
- 10.Kryzhanovsky, B.V., Litinskii, L.B., Fonarev, A.: An effective associative memory for pattern recognizing. In: 5-th Int. sympos. on Idvances in Inteligent Data Analisis 2003, Germany, pp. 179–187. Springer, Berlin (2003)Google Scholar
- 11.Mikaelyan, A.L., Kryzhanovsky, B.V., Litinskii, L.B.: Parametrical Neural Network. Optical Memory & Neural Network 12(3), 227–236 (2003)Google Scholar
- 12.Kryzhanovsky, B.V., Litinskii, L.B., Mikaelyan, A.L.: Vector-neuron models of associative memory. In: Proc. of Int. Joint Conference on Neural Networks IJCNN 2004, Budapest 2004, pp. 909–1004 (2004)Google Scholar
- 13.Kryzhanovsky, B.V., Mikaelyan, A.L., Fonarev, A.B.: Vector Neural Net Identifing Many Strongly Distorted and Correlated Patterns. In: Int. conf on Information Optics and Photonics Technology, Photonics Asia-2004, Beijing-2004. Proc. of SPIE, vol. 5642, pp. 124–133 (2004)Google Scholar
- 14.Alieva, D.I., Kryzhanovsky, B.V., Kryzhanovsky, V.M., Fonarev, A.B.: Q-valued neural network as a system of fast identification and pattern recognizing. Pattern Recognizing and Image Analysis 15(1), 30–33 (2005)Google Scholar
- 15.Kryzhanovsky, B.V., Kryzhanovsky, V.M., Mikaelian, A.L., Fonarev, A.B.: Parametrical Neural Network for Binary Patterns Identification. Optical Memory & Neural Network 14(2), 81–90 (2005)Google Scholar
- 16.Kryzhanovsky, B.V., Kryzhanovsky, V.M., Fonarev, A.B.: Decorrelating Parametrical Neural Network. In: Proc. of IJCNN Montreal, pp. 1023–1026 (2005)Google Scholar