Abstract
The quite large standard class of additive neural networks is considered from the point of view of the qualitative theory of differential equations. Connections with the theory of absolute stability are pointed out and a new class of Liapunov functions is introduced, starting from the positiveness theory (Yakubovich-Kalman-Popov lemma). The results are valid for a quite large class of dynamical systems and they are tested on some neural network structures. In the concluding part some perspective research is mentioned, including synchronization and time-delay effects.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Barbălat, I., Halanay, A.: Conditions de comportement “presque linéaire” dans la théorie des oscillations. Rev. Roum. Sci. Techn.-Electrotechn. et Energ. 19(2), 321–341 (1974)
Calvert, B.D., Marinov, C.A.: Another K-Winners-Take-All analog neural network. IEEE Trans. Neural Networks 11(4), 829–838 (2000)
Cohen, M.A.: The construction of arbitrary stable dynamics in nonlinear neural networks. Neural Networks 5, 83–103 (1992)
Danciu, D.: Dynamics of neural networks as nonlinear systems with several equilibria. In: Pazos, A., Sierra, A., Buceta, W. (eds.) Advancing Artificial Intelligence through Biological Process Applications, pp. 331–357. Medical Information Science Reference, IGI Global (2009)
Danciu, D., Răsvan, V.: On Popov-type stability criteria for neural networks. In: Elec. J. Qual. Theory Diff. Equ (Proc. 6th Coll. Qual. Theory Diff. Equ. QTDE), pp. 1–10 (1999), http://www.math.u-szeged.hu/ejqtde/6/623.pdf
Danciu, D., Răsvan, V.: Dynamics of neural networks - some qualitative properties. In: Sandoval, F., Prieto, A.G., Cabestany, J., Graña, M. (eds.) IWANN 2007. LNCS, vol. 4507, pp. 8–15. Springer, Heidelberg (2007)
Danciu, D., Răsvan, V.: Neural networks. equilibria, synchronization, delays. In: Dopico, J.R., Dorado, J., Pazos, A. (eds.) Encyclopedia of Artificial Intelligence, pp. 1219–1225. Information Science Reference, IGI Global (2009)
Danciu, D., Răsvan, V.: Gradient like behavior and high gain design of KWTA neural networks. In: Cabestany, J., Sandoval, F., Prieto, A., Corchado, J.M. (eds.) IWANN 2009. LNCS, vol. 5517, pp. 24–32. Springer, Heidelberg (2009)
Forti, M.: Some extensions of a new method to analyze complete stability of neural networks. IEEE Trans. Neural Networks 13(5), 1230–1238 (2002)
Gantmakher, F.R.: Theory of matrices, fourth augmented edn.“Nauka” Publishing House, Moscow (1988) (in Russian)
Ghelig, A.K., Leonov, G.A., Yakubovich, V.A.: Stability of Stationary Sets in Control Systems with Discontinuous Nonlinearities. World Scientific, Singapore (2004)
Gupta, M.M., Jin, L., Homma, N.: Dynamic Neural Networks. IEEE Press & J. Wiley, N. Y (2003)
Haddad, W.M.: Correction to “absolute stability criteria for multiple slope-restricted monotonic nonlinearities”. IEEE Trans. on Autom. Control 42(4), 591 (1997)
Haddad, W.M., Kapila, V.: Absolute stability criteria for multiple slope-restricted monotonic nonlinearities. IEEE Trans. on Autom. Control 40(2), 361–365 (1995)
Halanay, A., Răsvan, V.: Absolute stability of feedback systems with several differentiable non-linearities. Int. J. Systems Sci. 22(10), 1911–1927 (1991)
Halanay, A., Răsvan, V.: Applications of Liapunov Methods to Stability, Mathematics and its Applications, vol. 245. Kluwer Academic Publishers, Dordrecht (1993)
Leonov, G.A., Reitmann, V., Smirnova, V.B.: Non-local methods for pendulum-like feedback systems, Teubner Texte zur Mathematik, vol. 132. Teubner Verlag, Stuttgart-Leipzig (1992)
Noldus, E., Loccufier, M.: An application of Liapunov’s method for the analysis of neural networks. Journ. of Comp.and Appl. Math. 50, 425–432 (1995)
Noldus, E., Vingerhoeds, R., Loccufier, M.: Stability of analogue neural classification networks. Int. Journ. Systems. Sci. 25(1), 19–31 (1994)
Popov, V.M.: Hyperstability of Control Systems. Editura Academiei & Springer-Verlag, Bucharest & Heidelberg (1973)
Răsvan, V.: Dynamical systems with several equilibria and natural Liapunov functions. Archivum mathematicum 34(1), 207–215 (1998)
Răsvan, V., Danciu, D., Popescu, D.: On absolute (robust) stability: slope restrictions and stability multipliers. Int. J. Rob. Nonlin. Contr. (2011) (submitted)
Yakubovich, V.A.: Frequency domain conditions of absolute stability and dissipativeness of control systems with a single differentiable element (in Russian). Doklady Akad. Nauk SSSR 160(2), 298–301 (1965)
Yakubovich, V.A.: The method of the matrix inequalities in the theory of stability for nonlinear controlled systems II. Absolute stability in the class of slope restricted nonlinearities (in Russian). Avtom. i telemekh. 29(4), 588–590 (1965)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Danciu, D., Răsvan, V. (2011). Systems with Slope Restricted Nonlinearities and Neural Networks Dynamics. In: Cabestany, J., Rojas, I., Joya, G. (eds) Advances in Computational Intelligence. IWANN 2011. Lecture Notes in Computer Science, vol 6692. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-21498-1_71
Download citation
DOI: https://doi.org/10.1007/978-3-642-21498-1_71
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-21497-4
Online ISBN: 978-3-642-21498-1
eBook Packages: Computer ScienceComputer Science (R0)