Abstract
The objective of this research is the numerical estimation of the roots of a complete 2 × 2 nonlinear algebraic system of polynomial equations using a feed forward back-propagation neural network. The main advantage of this approach is the simple solution of the system, by building a structure—including product units—that simulates exactly the nonlinear system under consideration and find its roots via the classical back-propagation approach. Examples of systems with four or multiple roots were used, in order to test the speed of convergence and the accuracy of the training algorithm. Experimental results produced by the network were compared with their theoretical values.
Similar content being viewed by others
Notes
The originality of this idea has to be attributed to Dr. Miltiades Adamopoulos, an Emeritus Professor of the Technological Educational Institute of Thessaloniki.
References
Press W, Teukolsky S, Vetterling W, Flannery B (1992) Numerical recipes in C—the art of scientific programming, 2nd edn. Cambridge University Press, Cambridge
Abaffy J, Broyden CG, Spedicato E (1984) A class of direct methods for linear systems. Numerische Mathematik 45:361–376
Abaffy J, Spedicato E (1989) ABS projection algorithms: mathematical techniques for linear and nonlinear equations. Ellis Horwood, Chichester
Spedicato E, Bodon E, Del Popolo A, Mahdavi-Amiri N (2000) ABS methods and ABSPACK for linear systems and optimization, a review. In: Proceedings of the 3rd seminar of numerical analysis, Zahedan, November 15/17, University of Zahedan
Abaffy J, Galantai A (1987) Conjugate direction methods for linear and nonlinear systems of algebraic equations. Colloquia Mathematica Soc. Jfinos Bolyai. Numerical Methods, Miskolc, Hungary, 1986; Edited by R6zsa P and Greenspan D, North Holland, Amsterdam, Netherlands, vol 50, pp 481–502
Abaffy J, Galantai A, Spedicato E (1987) The local convergence of ABS methods for nonlinear algebraic equations. Numerische Mathematik 51:429–439
Galantai A, Jeney A (1996) Quasi-Newton ABS methods for solving nonlinear algebraic systems of equations. J Opt Theory Appl 89(3):561–573
Kublanovskaya VN, Simonova VN (1996) An approach to solving nonlinear algebraic systems. 2. J Math Sci 79(3):1077–1092
Emiris I, Mourrain B, Vrahatis M (1999) Sign methods for counting and computing real roots of algebraic systems. Technical Report RR-3669. INRIA, Sophia Antipolis
Mejzlik P (1994) A bisection method to find all solutions of a system of nonlinear equations. Domain decomposition methods in scientific and engineering computing. Book Chapter, AMS
Nakaya Y, Oishi S (1998) Find all solutions of nonlinear systems of equations using linear programming with guaranteed accuracies. J Univ Comput Sci 4(2):171–177
Grosan C, Abraham A (1996) Solving nonlinear equation systems using evolutionary algorithms. In: Proceedings of genetic & evolutionary computation conference, Seattle, USA, Proceedings on CD
Chistine J, Ralz D, Nerep K, Pau\(\Uppi\) (March 1995) A combined method for enclosing all solutions of nonlinear systems of polynomial equations. Reliable computing. Springer, Berlin 1(1):41–64
Tsai T-F, Lin M-H (2007) Finding all solutions of systems of nonlinear equations with free variables. Eng Opt 39(6):649–659
Michael WS, Chun C (2001) An algorithm for finding all solutions of a nonlinear system. J Comput Appl Math 137(2):293–315
Xue J, Zi Y, Feng Y, Yang L, Lin Z (2004) An intelligent hybrid algorithm for solving nonlinear polynomial systems. In: Bubak M et al (eds) Proceedings of international conference on computational science, LCNS 3037, pp 26–33
Dolotin V, Morozov A (September 2006) Introduction to nonlinear algebra v.2. online document found in the Arxiv repository (http://www.arxiv.org), ITEP, Moscow, Russia
Zhang G, Bi L, Existence of solutions for a nonlinear algebraic system. Discrete dynamics in nature and society, Volume 2009, Article ID 785068, available online from the URL http://www.hindawi.com/journals/ddns/2009/785068.html
Margaritis KG, Adamopoulos M, Goulianas K (1993) Solving linear systems by artificial neural network energy minimisation. University of Macedonia Annals, vol XII, pp 502–525, (in Greek)
Margaritis KG, Adamopoulos M, Goulianas K, Evans DJ (1994) Artificial neural networks and iterative linear algebra methods. Parallel Algorithms Appl 3(1–2):31–44
Mathia K, Saeks R (1995) Solving nonlinear equations using recurrent neural networks. World congress on neural networks, July 17–21, Washington DC, USA
Mishra D, Kalva PK, Modified Hopfield neural network approach for solving nonlinear algebraic equations. Engineering Letters, 14:1, EL_14_01_23 (advance online publication: 12 February 2007)
Luo D, Han Z (1995) Solving nonlinear equation systems by networks. In: Proceedings of international conference on systems, man and cybernetics, vol 1, pp 858–862
Li Guimei, Zhezhao Z (2008) A neural network algorithm for solving nonlinear equation systems. In: Proceedings of 20th international conference on computational intelligence and security, pp 20–23, Los Alamito, USA
Margaris A, Adamopoulos M (2007) Solving nonlinear algebraic systems using artificial neural networks. In: Proceedings of the 10th international conference on engineering applications of artificial neural networks, August, Thessaloniki, Greece
Margaris A, Adamopoulos M (2009) Identifying fixed points of Henon map using artificial neural networks. In: Proceedings of the 2nd chaotic modeling and simulation international conference, CHAOS2009, June 1–5, Chania, Crete, Greece, pp 107–120
Freeman JA, Skapura DM (1999) Neural networks: algorithms, applications and programming techniques. Addison Wesley: ISBN 0-201-51376-5
Ko KH, Sakkalis T, Patrikalakis NM (2004) Nonlinear polynomial systems: multiple roots and their multiplicities. In: Giannini F, Pasko A (eds) Proceedings of shape modelling international conference, SMI 2004, Genova, Italy
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Margaris, A., Goulianas, K. Finding all roots of 2 × 2 nonlinear algebraic systems using back-propagation neural networks. Neural Comput & Applic 21, 891–904 (2012). https://doi.org/10.1007/s00521-010-0488-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-010-0488-z