×

Hopf bifurcation in bidirectional associative memory neural networks with delays: Analysis and computation. (English) Zbl 1054.65076

This paper is concerned with the stability behaviour of some delay differential equations that model bidirectional associative memory neural networks. The system under consideration has the form \(x_i'(t)=-x_i(t)+\sum_{j=1}^na_{ij}f_j(y_j(t-\tau_1))\), \(y_i'(t)=-y_i(t)+\sum_{j=1}^nb_{ij}g_j(y_j(t-\tau_2))\), \(i=1,\dots,N\), where \(a_{ij},b_{ij}\) are the constant connection weights through the neurons in two layers, \(f_i\) and \(g_i\) given activation functions and \(\tau_1,\tau_2\) constant delays.
In this problem several sets of sufficient conditions on the data problem \(a_{ij},b_{ij},f_i,g_i\) that imply the asymptotic stability of the zero solution are given. Also the existence of Hopf bifurcations in terms of \(\tau=\tau_1+\tau_2\) is studied proving that for some \(\tau=\tau^*\) the above system has a branch of periodic solutions bifurcating from the zero solution near \(\tau=\tau^*\). Finally, a particular system with \(N=2\) neurons is considered to test the above theoretical results using the delay differential equations solver of L. F. Shampine and S. Thompson [Appl. Numer. Math. 37, 441–458 (2001; Zbl 0983.65079)].

MSC:

65L07 Numerical investigation of stability of solutions to ordinary differential equations
34K20 Stability theory of functional-differential equations
34K18 Bifurcation theory of functional-differential equations
92B20 Neural networks for/in biological studies, artificial life and related topics
34K25 Asymptotic theory of functional-differential equations

Citations:

Zbl 0983.65079

Software:

dde23
PDF BibTeX XML Cite
Full Text: DOI

References:

[1] Bélair, J., Stability in a model of a delayed neural network, J. dyn. differential equations, 5, 607-623, (1993) · Zbl 0796.34063
[2] Campbell, S.A.; Ruan, S.; Wei, J., Qualitative analysis of a neural network model with multiple time delays, Internat. J. bifur. chaos appl. sci. eng., 9, 8, 1585-1595, (1999) · Zbl 1192.37115
[3] Cao, Y.; Wu, Q., A note on stability of analog neural networks with time delays, IEEE trans. neural networks, 7, 1533-1535, (1996)
[4] Cooke, K.; van den Driessche, P., On zeros of some transcendental equation, Funkcial. ekvac., 29, 77-90, (1985)
[5] Deimling, K., Nonlinear functional analysis, (1985), Springer New York · Zbl 0559.47040
[6] Feng, C.; Plamondon, R., On the stability analysis of delayed neural networks systems, Neural networks, 14, 1181-1188, (2000)
[7] Fielder, M., Special matrices and their applications in numerical mathematics, (1986), Martinus Nijhoff Publishers Boston
[8] Forti, M., On global asymptotic stability of a class of nonlinear systems arising in neural network theory, J. differential equations, 113, 246-264, (1994) · Zbl 0828.34039
[9] Gopalsamy, K.; He, X., Delay-independent stability in bidirectional associative memory networks, IEEE trans. neural networks, 5, 998-1002, (1994)
[10] Gopalsamy, K.; He, X., Stability in asymmetric Hopfield nets with transmission delays, Phys. D, 76, 344-358, (1994) · Zbl 0815.92001
[11] Guan, Z.; Chen, G.; Qin, Y., On equilibria, stability, and instability of Hopfield neural networks, IEEE trans. neural networks, 2, 534-540, (2000)
[12] Hassard, B.D.; Kazavinoff, N.D.; Wan, Y.H., Theory and applications of the Hopf bifurcation, (1981), Cambridge University Press Cambridge, MA · Zbl 0474.34002
[13] Hopfield, J., Neurons with graded response have collective computational properties like those of two-stage neurons, Proc. natl. acad. sci. USA, 81, 3088-3092, (1984) · Zbl 1371.92015
[14] Kazarinoff, N.D.; Wan, Y.H.; van den Driessche, P., Hopf bifurcation and stability of periodic solutions of differential-difference and integro-differential equations, J. inst. math. appl., 21, 461-477, (1978) · Zbl 0379.45021
[15] Kosko, B., Adaptive bidirectional associative memories, Appl. opt., 26, 4947-4960, (1987)
[16] Kosko, B., Bidirectional associative memories, IEEE trans. systems man cybernet., 18, 49-60, (1988)
[17] Kosko, B., Unsupervised learning in noise, IEEE trans. neural networks, 1, 1-12, (1990)
[18] Lu, H., On stability of nonlinear continuous-time neural networks with delays, Neural networks, 13, 1135-1143, (2000)
[19] Marcus, C.; Westervelt, R., Stability of analog neural networks with delay, Phys. rev. A, 39, 347-359, (1989)
[20] Matsuoka, K., Stability conditions for nonlinear continuous neural networks with asymmetric connection weights, Neural networks, 5, 495-500, (1992)
[21] Mohamad, S., Globaly exponential stability in continuous-time and discrete-time delayed bidirectional neural networks, Physica D, 159, 233-251, (2001) · Zbl 0984.92502
[22] Shampine, L.F.; Thompson, S., Solving DDEs in MATLAB, Appl. numer. math., 37, 441-458, (2001) · Zbl 0983.65079
[23] Shayer, L.P.; Campbell, S.A., Stability, bifurcation, and multistability in a system of two coupled neurons with multiple time delays, SIAM J. appl. math., 61, 2, 673-700, (2000) · Zbl 0992.92013
[24] Tank, D.; Hopfield, J., Simple neural optimization networksan A/D converter, signal decision circuit and a linear programming circuit, IEEE trans. circuits systems, 33, 533-541, (1986)
[25] van den Driessche, P.; Zou, X., Global attractivity in delayed Hopfield neural network models, SIAM J. appl. math., 58, 1878-1890, (1998) · Zbl 0917.34036
[26] Wei, J.; Ruan, S., Stability and bifurcation in a neural network model with two delays, Phys. D, 130, 255-272, (1999) · Zbl 1066.34511
[27] Wu, J., Symmetric functional-differential equations and neural networks with memory, Trans. amer. math. soc., 350, 4799-4838, (1999) · Zbl 0905.34034
[28] Wu, J.; Zou, X., Patterns of sustained oscillations in neural networks with delayed iteractions, Appl. math. comput., 73, 55-75, (1995) · Zbl 0857.92003
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.