Abstract
Using a probabilistic approach, the parallel dynamics of fully connected Q-Ising neural networks is studied for arbitrary Q. A novel recursive scheme is set up to determine the time evolution of the order parameters through the evolution of the distribution of the local field, taking into account all feedback correlations. In contrast to extremely diluted and layered network architectures, the local field is no longer normally distributed but contains a discrete part. As an illustrative example, an explicit analysis is carried out for the first four time steps. For the case of the Q = 2 and Q = 3 model the results are compared with extensive numerical simulations and excellent agreement is found. Finally, equilibrium fixed-point equations are derived and compared with the thermodynamic approach based upon the replica-symmetric mean-field approximation.
Similar content being viewed by others
REFERENCES
B. Derrida, E. Gardner, and A. Zippelius, Europhys. Lett. 4:167 (1987).
D. Bollé, G. M. Shim, B. Vinck, and V. A. Zagrebnov, J. Stat. Phys. 74:565 (1994).
E. Domany, W. Kinzel, and R. Meir, J. Phys. A: Math. Gen. 22:2081 (1989).
D. Bollé, G. M. Shim, and B. Vinck, J. Stat. Phys. 74:583 (1994).
T. L. H. Watkin and D. Sherrington, J. Phys. A: Math. Gen. 24:5427 (1991).
A. E. Patrick and V. A. Zagrebnov, J. Phys. A: Math. Gen. 23:L1323 (1990); 25:1009 (1992).
W. Kinzel, Z. Phys. B 60:205 (1985).
E. Gardner, B. Derrida, and P. Mottishaw, J. Physique 48:741 (1987).
S. Amari and K. Maginu, Neural Networks 1:63 (1988).
W. Krauth, J. P. Nadal, and M. Mezard, J. Phys. A: Math. Gen. 21:2995 (1988).
H. Horner, D. Bormann, M. Frick, H. Kinzelbach, and A. Schmidt, Z. Phys. B 76:381 (1989).
R. D. Henkel and M. Opper, Europhys. Lett. 11:403 (1990); J. Phys. A: Math. Gen. 24:2201 (1991).
A. E. Patrick and V. A. Zagrebnov, J. Stat. Phys. 63:59 (1991).
A. E. Patrick and V. A. Zagrebnov, J. Phys. A: Math. Gen. 24:3413 (1991).
M. Okada, Neural Networks 9:1429 (1996).
D. Bollé, B. Vinck, and V. A. Zagrebnov, J. Stat. Phys. 70:1099 (1993).
P. Peretto, Biol. Cybern. 50:51 (1984).
J. L. van Hemmen and R. Kühn, in Models of Neural Networks, E. Domany, J. L. van Hemmen, and K. Schulten, eds. (Springer, 1991), p. 1.
J. F. Fontanari and Köberle, J. Physique 49:13 (1988).
T. Stiefvater, K. R. Müller, and R. Kühn, Physica A 232:61 (1996).
D. Bollé, H. Rieger, and G. M. Shim, J. Phys. A: Math. Gen. 27:3411 (1994).
E. Barkai, I. Kanter, and H. Sompolinsky, Phys. Rev. A 41:590 (1990).
A. N. Shiryayev, Probability (Springer, New York, 1984).
M. Shiino and T. Fukai, J. Phys. A: Math. Gen. 25:L375 (1992).
M. Shiino and T. Fukai, Phys. Rev. E 48:867 (1993).
D. Amit, H. Gutfreund, and H. Sompolinsky, Ann. Phys. (N.Y.) 173:30 (1987).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Bollé, D., Jongen, G. & Shim, G.M. Parallel Dynamics of Fully Connected Q-Ising Neural Networks. Journal of Statistical Physics 91, 125–153 (1998). https://doi.org/10.1023/A:1023088004195
Issue Date:
DOI: https://doi.org/10.1023/A:1023088004195