Publication: A continuous Hopfield network equilibrium points algorithm
Loading...
Full text at PDC
Publication Date
2005-08
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Pergamon-Elsevier Science Ltd
Citations
Exportar
Abstract
The continuous Hopfield network (CHN) is a classical neural network model. It can be used to solve some classification and optimization problems in the sense that the equilibrium points of a differential equation system associated to the CHN is the solution to those problems. The Euler method is the most widespread algorithm to obtain these CHN equilibrium points, since it is the simplest and quickest method to simulate complex differential equation systems. However, this method is highly sensitive with respect to initial conditions and it requires a lot of CPU time for medium or greater size CHN instances. In order to avoid these shortcomings, a new algorithm which obtains one equilibrium point for the CHN is introduced in this paper. It is a variable time-step method with the property that the convergence time is shortened; moreover, its robustness with respect to initial conditions will be proven and some computational experiences will be shown in order to compare it with the Euler method
Description
UCM subjects
Unesco subjects
Keywords
Citation
Aiyer SVB, Niranjan M, Fallside F. A theoretical investigation into the performance of the Hofield model. IEEE Transactions on Neural Networks 1990;1(2):204–15.
Hopfield JJ. Neurons with graded response have collective computational properties like those of two-states neurons. Proceedings of the National Academy of Sciences USA 1984;81:3088–92.
Ghosh A, Pal NR, Pal SK. Object background classi%cation using Hopfield type neural networks. International Journal of Pattern Recognition and Artificial Intelligence 1992;6:989–1008.
Nasrabadi NM, Choo CY. Hopfield network for stereo vision correspondence. IEEE Transactions on Neural Network 1992;3(1):5–13.
Wasserman PD. Neural computing. Theory and practice. New York: Van Nostrand Reinhold; 1989.
Wu JK. Neural networks and simulation methods. New York: Marcel Dekker; 1994.
Hopfield JJ, Tank DW. Neural computation of decisions in optimization problems. Biological Cybernetics 1985;52: 1–25.
Talaván PM, Yáñez J. Parameter setting of the Hopfield network applied to TSP. Neural Networks 2002;15(3): 363–73.
Talaván PM. El modelo de Hopfield aplicado a problemas de optimizacion combinatoria. PhD dissertation. Universidad Complutense de Madrid, Spain; 2003.
Demidowitsch BP, Maron IA, Schuwalowa ES. M-etodos numericos de Analisis. Madrid: Paraninfo; 1980.
Bagherzadeh N, Kerola T, Leddy B, Brice R. On parallel execution of the traveling salesman problem on a neural network model. Proceedings of the IEEE International Conference on Neural Networks, San Diego, USA, vol. III, 1987. p. 317–24.
Brandt RD, Wang Y, Laub AJ, Mitra SK. Alternative networks for solving the traveling salesman problem and the list-matching problem. Proceedings of the International Conference on Neural Network (ICNN’88), San Diego, USA, vol. II, 1988. p. 333–40.
Abe S. Global convergence and suppression of spurious states of the Hopfield neural networks. IEEE Transactions on Circuits and Systems 1993;40(4):246–57.
Wang L, Smith K. On chaotic simulated annealing. IEEE Transactions on Neural Network 1998;9(4):716–8.