Document Zbl 1438.34238

Akhmet, Marat; Aruğaslan Çinçin, Duygu; Cengiz, Nur

Exponential stability of periodic solutions of recurrent neural networks with functional dependence on piecewise constant argument. (English) Zbl 1438.34238

Turk. J. Math. 42, No. 1, 272-292 (2018).

Summary: In this study, we develop a model of recurrent neural networks with functional dependence on piecewise constant argument of generalized type. Using the theoretical results obtained for functional differential equations with piecewise constant argument, we investigate conditions for existence and uniqueness of solutions, bounded solutions, and exponential stability of periodic solutions. We provide conditions based on the parameters of the model.

Cited in 3 Documents

MSC:

34K13	Periodic solutions to functional-differential equations
34K20	Stability theory of functional-differential equations
92B20	Neural networks for/in biological studies, artificial life and related topics

Keywords:

recurrent neural networks; functional differential equations; piecewise constant argument; periodic solution; stability

Cite Review PDF

Full Text: DOI

References:

[1]	Aftabizadeh AR, Wiener J, Xu JM. Oscillatory and periodic solutions of delay diﬀerential equations with piecewise constant argument. P Am Math Soc 1987; 99: 673-679. · Zbl 0631.34078 · doi:10.1090/S0002-9939-1987-0877038-7
[2]	Akhmet MU. On the integral manifolds of the diﬀerential equations with piecewise constant argument of generalized type. In: Agarval RP, Perera K, editors. Proceedings of the Conference on Diﬀerential and Diﬀerence Equations at the Florida Institute of Technology; 1-5 August 2005; Melbourne, Florida. Cairo, Egypt: Hindawi Publishing Corp., 2006, pp. 11-20. · Zbl 1133.34040
[3]	Akhmet MU. Stability of diﬀerential equations with piecewise constant arguments of generalized type. Nonlinear Anal 2008; 68: 794-803. · Zbl 1173.34042 · doi:10.1016/j.na.2006.11.037
[4]	Akhmet MU. Nonlinear Hybrid Continuous Discrete Time Models. Amsterdam, the Netherlands: Atlantis Press, 2011. · Zbl 1328.93001 · doi:10.2991/978-94-91216-03-9
[5]	Akhmet MU. Quasilinear retarded diﬀerential equations with functional dependence on piecewise constant argument. Commun Pure Appl Ana 2014; 13: 929-947. · Zbl 1285.34059 · doi:10.3934/cpaa.2014.13.929
[6]	Akhmet MU, Aru˘gaslan D. Lyapunov-Razumikhin method for diﬀerential equations with piecewise constant argument. Discret Contin Dyn-A 2009; 25: 457-466. · Zbl 1179.34077 · doi:10.3934/dcds.2009.25.457
[7]	Akhmet MU, Aru˘gaslan D, Yılmaz E. Stability analysis of recurrent neural networks with piecewise constant argument of generalized type. Neural Networks 2010; 23: 805-811. · Zbl 1400.34112 · doi:10.1016/j.neunet.2010.05.006
[8]	Akhmet MU, Aru˘gaslan D, Yılmaz E. Stability in cellular neural networks with a piecewise constant argument. J Comput Appl Math 2010; 233: 2365-2373. · Zbl 1191.68484 · doi:10.1016/j.cam.2009.10.021
[9]	Akhmet MU, Yılmaz E. Hopfield-type neural networks systems equations with piecewise constant argument. Int J Qual Theory Diﬀer Equat Appl 2009; 3: 8-14. · Zbl 1263.34019
[10]	Akhmet MU, Yılmaz E. Global attractivity in impulsive neural networks with piecewise constant delay. In: Proceedings of Neural, Parallel, and Scientific Computations. Atlanta, GA, USA: Dynamic Publishers, Inc., 2010, pp. 11-18. · Zbl 1221.92004
[11]	Akhmet MU, Yılmaz E. Global exponential stability of neural networks with non-smooth and impact activations. Neural Networks 2012; 34: 18-27. · Zbl 1258.34157 · doi:10.1016/j.neunet.2012.06.004
[12]	Akhmet MU, Yılmaz E. Neural Networks with Discontinuous/Impact Activations. New York, NY, USA: Springer, 2014. · Zbl 1295.92007 · doi:10.1007/978-1-4614-8566-7
[13]	Aru˘gaslan D. Diﬀerential equations with discontinuities and population dynamics. PhD, Middle East Technical University, Ankara, Turkey, 2009.
[14]	Belair J, Campbell SA, Driessche PVD. Frustration, stability, and delay-induced oscillations in a neural network model. SIAM J Appl Math 1996; 56: 245-255. · Zbl 0840.92003 · doi:10.1137/S0036139994274526
[15]	Chen TP. Global exponential stability of delayed Hopfield neural networks. Neural Networks 2001; 14: 977-980. · doi:10.1016/S0893-6080(01)00059-4
[16]	Cooke KL, Wiener J. Retarded diﬀerential equations with piecewise constant delays. J Math Anal Appl 1984; 99: 265-297. · Zbl 0557.34059 · doi:10.1016/0022-247X(84)90248-8
[17]	Cooke KL, Wiener J. Stability for linear equations with piecewise continuous delay. Comput Math Appl-A 1986; 12: 695-701. · Zbl 0609.34080 · doi:10.1016/0898-1221(86)90055-6
[18]	Cooke KL, Wiener J. An equation alternately of retarded and advanced type. P Am Math Soc 1987; 99: 726-732. · Zbl 0628.34074 · doi:10.1090/S0002-9939-1987-0877047-8
[19]	Cooke KL, Wiener J. Neutral diﬀerential equations with piecewise constant argument. Boll Un Mat Ital 1987; 7: 321-346. · Zbl 0639.34003
[20]	Driessche PVD, Zou X. Global attractivity in delayed Hopfield neural network models. SIAM J Appl Math 1998; 58: 1878-1890. · Zbl 0917.34036 · doi:10.1137/S0036139997321219
[21]	Hale J. Functional Diﬀerential Equations. New York, NY, USA: Springer, 1971. · Zbl 0222.34003 · doi:10.1007/978-1-4615-9968-5
[22]	Papaschinopoulos G. On the integral manifold for a system of diﬀerential equations with piecewise constant argument. J Math Anal Appl 1996; 201: 75-90. · Zbl 0853.34059 · doi:10.1006/jmaa.1996.0242
[23]	Wiener J. Generalized Solutions of Functional Diﬀerential Equations. Singapore: World Scientific, 1993. · Zbl 0874.34054 · doi:10.1142/1860
[24]	Wiener J, Lakshmikantham V. A damped oscillator with piecewise constant time delay. Nonlinear Stud 2000; 7: 78-84. · Zbl 1016.34069
[25]	Yang X. Existence and exponential stability of almost periodic solution for cellular neural networks with piecewise constant argument. Acta Math Appl Sin 2006; 29: 789-800.

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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.