Abstract
A novel technique is presented that implements continuous-state Hopfield neural networks on a digital computer. Instead of the usual forward Euler rule, the backward method is used. The stability and Lyapunov function of the proposed discrete model are indirectly guaranteed, even for reasonably large step size. This is possible because discretization by implicit numerical methods inherits the stability of the continuous-time model. On the contrary, the forward Euler method requires a very small step size to guarantee convergence to solutions. The presented technique takes advantage of the extensive research on continuous-time stability, as well as recent results in the field of dynamical analysis of numerical methods. Also, standard numerical methods allow for synchronous activation of neurons, thus leading to performance enhancement. Numerical results are presented that illustrate the validity of this approach when applied to optimization problems.
This work has been partially supported by the Spanish Ministerio de Ciencia y Tecnología (MCYT), Project No. TIC2001-1758.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Hopfield, J.: Neural networks and physical systems with emergent collective computational abilities. Proc. Natl. Acad. Sci. USA 79 (1982) 2554–2558
Hopfield, J.: Neurons with graded response have collective computational properties like those of two-state neurons. Proc. Natl. Acad. Sci. USA 81 (1984) 3088–3092
Tank, D., Hopfield, J.: ‘Neural’ computation of decisions in optimization problems. Biol. Cybern. 52 (1985) 141–152
Joya, G., Atencia, M.A., Sandoval, F.: Hopfield neural networks for optimization: Study of the different dynamics. Neurocomputing 43 (2002) 219–237
Vidyasagar, M.: Are analog neural networks better than binary neural networks? Circuits, Systems and Signal Processing 17 (1998) 243–270
Vidyasagar, M.: Minimum-seeking properties of analog neural networks with multilinear objective functions. IEEE Trans. On Automatic Control 40 (1995) 1359–1375
Chen, T., Amari, S.I.: New theorems on global convergence of some dynamical systems. Neural Networks 14 (2001) 251–255
Smith, K., Palaniswami, M., Krishnamoorthy, M.: Neural techniques for combinatorial optimization with applications. IEEE Trans. On Neural Networks 9 (1998) 1301–1318
Bharitkar, S., Tsuchiya, K., Takefuji, Y.: Microcode optimization with neural networks. IEEE Transactions on Neural Networks 10 (1999) 698–703
Wang, L.: On the dynamics of discrete-time, continuous-state Hopfield neural networks. IEEE Trans. On Circuits and Systems-II 45 (1998) 747–749
Tino, P., Horne, B., Giles, C.: Attractive periodic sets in discrete-time recurrent networks (with emphasis on fixed-point stability and bifurcations in two-neuron networks). Neural Computation 13 (2001) 1379–1414
Stuart, A., Humphries, A.: Dynamical systems and numerical analysis. Cambridge University Press (1996)
Abe, S.: Theories on the Hopfield neural networks. In: Proc. IEE International Joint Conference on Neural Networks. Volume I. (1989) 557–564
Galan-Marin, G., Muñoz Perez, J.: Design and analysis of maximum Hopfield networks. IEEE Trans. On Neural Networks 12 (2001) 329–339
Atencia, M.A., Joya, G., Sandoval, F.: Numerical implementation of continuous Hopfield networks for optimization. In: Proc. European Symposium on Artificial Neural Networks. (2001) 359–364
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Atencia, M.A., Joya, G., Sandoval, F. (2002). Continuous-State Hopfield Dynamics Based on Implicit Numerical Methods. In: Dorronsoro, J.R. (eds) Artificial Neural Networks — ICANN 2002. ICANN 2002. Lecture Notes in Computer Science, vol 2415. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46084-5_220
Download citation
DOI: https://doi.org/10.1007/3-540-46084-5_220
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44074-1
Online ISBN: 978-3-540-46084-8
eBook Packages: Springer Book Archive