Previous |  Up |  Next


non-convex quadratic optimization; recurrent neural network model; global optimality conditions; global convergence
In this paper, based on a generalized Karush-Kuhn-Tucker (KKT) method a modified recurrent neural network model for a class of non-convex quadratic programming problems involving a so-called $Z$-matrix is proposed. The basic idea is to express the optimality condition as a mixed nonlinear complementarity problem. Then one may specify conditions for guaranteeing the global solutions of the original problem by using results from the S-lemma. This process is proved by building up a dynamic system from the optimality condition whose equilibrium point is exactly the solution of the mixed nonlinear complementarity problem. By the study of the resulting dynamic system it is shown that under given assumptions, steady states of the dynamic system are stable. Numerical simulations and comparisons with the other methods are presented to illustrate the efficiency of the practical technique that is proposed in this paper.
[1] Bazaraa, M. S., Shetty, C. M.: Nonlinear Programming Theory and Algorithms. Wiley and Sons, New York 1990. MR 0533477 | Zbl 1140.90040
[2] Beyer, D., Ogier, R.: Tabu learning: A neural network search method for solving nonconvex optimization problems. IEEE Int. Joint Conf. Neural Networks 2 (2000), 953-961.
[3] Bian, W., Xue, X.: Subgradient-based neural networks for nonsmooth nonconvex optimization problems. IEEE Trans. Neural Networks 20 (2009), 6, 1024-1038. DOI 10.1109/tnn.2009.2016340 | MR 2497796
[4] Chicone, C.: Ordinary Differential Equations with Applications. Second edition. Springer-Verlag, New York 2006. MR 2224508
[5] Forti, M., Nistri, P., Quincampoix, M.: Convergence of neural networks for programming problems via a nonsmooth Lojasiewicz inequality. IEEE Trans. Neural Networka 17 (2006), 6, 1471-1486. DOI 10.1109/tnn.2006.879775
[6] Gao, X. B.: A novel neural network for nonlinear convex programming problems. IEEE Trans. Neural Network 15 (2004), 613-621. DOI 10.1109/tnn.2004.824425
[7] Hu, X.: Neurodynamic optimization: Towards nonconvexity. In: Recurrent Neural Networks ( X. Hu and P. Balasubramaniam, ed.), IN-TECH, 2008, pp. 289-308. DOI 10.5772/5551
[8] Jeyakumar, V., Rubinov, A. M., Wu, Z. Y.: Non-convex quadratic minimization problems with quadratic constraints: global optimality conditions. Math. Program., Ser. A 110 (2007), 521-541. DOI 10.1007/s10107-006-0012-5 | MR 2324960 | Zbl 1206.90178
[9] Jeyakumar, V., Lee, G. M., Li, G. Y.: Alternative theorems for quadratic inequality systems and global quadratic optimization. SIAM J. Optim 20 (2009), 2, 983-1001. DOI 10.1137/080736090 | MR 2534772 | Zbl 1197.90315
[10] Jeyakumar, V., Srisatkunarajah, S.: Lagrange multiplier necessary condition for global optimality for non-convex minimization over a quadratic constraint via S-lemma. Optim. Lett. 3 (2009), 23-33. DOI 10.1007/s11590-008-0088-3 | MR 2453502
[11] Khalil, H. K.: Nonlinear Systems. Third edition. Prentice Hall, 2002.
[12] Malek, A.: Application of recurrent neural networks to optimization problems. In: Recurrent Neural Networks ( X. Hu and P. Balasubramaniam, eds.), IN-TECH, 2008, pp. 255-288. DOI 10.5772/5556
[13] Malek, A., Alipour, M.: Numerical solution for linear and quadratic programming problems using a recurrent neural network. Appl. Math. Comput 192 (2007), 27-39. DOI 10.1016/j.amc.2007.02.149 | MR 2385566 | Zbl 1193.90164
[14] Malek, A., Hosseinipour-Mahani, N., Ezazipour, S.: Efficient recurrent neural network model for the solution of general nonlinear optimization problems. Optimization Methods and Software 25 (2010), 489-506. DOI 10.1080/10556780902856743 | MR 2724153 | Zbl 1225.90129
[15] Malek, A., Ezazipour, S., Hosseinipour-Mahani, N.: Double projection neural network for solving pseudomonotone variational inequalities. Fixed Point Theory 12 (2011), 2, 401-418. MR 2895702
[16] Malek, A., Ezazipour, S., Hosseinipour-Mahani, N.: Projected dynamical systems and optimization problems. Bull. Iranian Math. Soc. 37 (2011), 2, 81-96. MR 2890580 | Zbl 1253.37091
[17] Malek, A., Yari, A.: Primal-dual solution for the linear programming problem using neural network. Appl. Math. Comput. 169 (2005), 198-211. DOI 10.1016/j.amc.2004.06.081 | MR 2170909
[18] Miller, R. K., Michel, A. N.: Ordinary Differential Equations. Academic Press, 1982. DOI 10.1016/b978-0-12-497280-3.50008-6 | MR 0660250 | Zbl 0552.34001
[19] Polik, I., Terlaky, T.: A survey of the S-Lemma. SIAM Rev. 49 (2007), 371-418. DOI 10.1137/s003614450444614x | MR 2353804 | Zbl 1128.90046
[20] Sun, C. Y., Feng, C. B.: Neural networks for nonconvex nonlinear programming problems: A switching control approach. In: Lecture Notes in Computer Science 3495, Springer-Verlag, Berlin 2005, pp. 694-699. DOI 10.1007/11427391_111 | Zbl 1082.68702
[21] Tao, Q., Liu, X., Xue, M. S.: A dynamic genetic algorithm based on continuous neural networks for a kind of non-convex optimization problems. Appl. Math. Comput. 150 (2004), 811-820. DOI 10.1016/s0096-3003(03)00309-6 | MR 2039677 | Zbl 1161.65333
[22] Tian, Y., Lu, Ch.: Nonconvex quadratic reformulations and solvable conditions for mixed integer quadratic programming problems. J. Industr. Managment Optim. 7 (2011), 1027-1039. DOI 10.3934/jimo.2011.7.1027 | MR 2824780 | Zbl 1231.90317
[23] Xia, Y., Feng, G., Wang, J.: A recurrent neural network with exponential convergence for solving convex quadratic program and related linear piecewise equation. Neural Networks 17 (2004), 1003-1015. DOI 10.1016/j.neunet.2004.05.006
[24] Xia, Y. S., Feng, G., Wang, J.: A novel recurrent neural network for solving nonlinear optimization problems with inequality constraints. IEEE Trans. Neural Networks 19 (2008), 1340-1353. DOI 10.1109/tnn.2008.2000273
[25] Xue, X., Bian, W.: A project neural network for solving degenerate convex quadratic program. Neurocomputing 70 (2007), 2449-2459. DOI 10.1016/j.neucom.2006.10.038
[26] Yan, Z., Wang, J., Li, G.: A collective neurodynamic optimization approach to bound-constrained nonconvex optimization. Neural networks 55 (2014), 20-29. DOI 10.1016/j.neunet.2014.03.006 | Zbl 1308.90137
[27] Yashtini, M., Malek, A.: Solving complementarity and variational inequalities problems using neural networks. Appl. Math. Comput. 190 (2007), 216-230. DOI 10.1016/j.amc.2007.01.036 | MR 2335442 | Zbl 1128.65052
[28] Zhang, Y.: Computing a Celis-Dennis-Tapia trust-region step for equality constrained optimization. Math. Programming 55 (1992), 109-124. DOI 10.1007/bf01581194 | MR 1163297 | Zbl 0773.90056
[29] Zheng, X. J., Sun, X. L., Li, D., Xu, Y. F.: On zero duality gap in nonconvex quadratic programming problems. Global Optim. 52 (2012), 229-242. DOI 10.1007/s10898-011-9660-y | MR 2886307 | Zbl 1266.90151
Partner of
EuDML logo