Abstract
In this paper, we propose a new method, namely the level-value estimation method, for finding global minimizer of continuous optimization problem. For this purpose, we define the variance function and the mean deviation function, both depend on a level value of the objective function to be minimized. These functions have some good properties when Newton’s method is used to solve a variance equation resulting by setting the variance function to zero. We prove that the largest root of the variance equation equals the global minimal value of the corresponding optimization problem. We also propose an implementable algorithm of the level-value estimation method where importance sampling is used to calculate integrals of the variance function and the mean deviation function. The main idea of the cross-entropy method is used to update the parameters of sample distribution at each iteration. The implementable level-value estimation method has been verified to satisfy the convergent conditions of the inexact Newton method for solving a single variable nonlinear equation. Thus, convergence is guaranteed. The numerical results indicate that the proposed method is applicable and efficient in solving global optimization problems.
Similar content being viewed by others
Notes
The three parameters H,V and B are used to give the upper bounds of CΔ, \(\operatorname{Var}(Y_{1})\) in (48) and the absolute value of lower bound of the objective function f over ℝn, respectively.
The scopes of parameters α,β and q are suggested in [23]. In the numerical tests of this paper, we set α=0.8,β=0.80 and q=6.
References
Pardalos, P.M., Romeijna, H., Tuy, H.: Recent developments and trends in global optimization. J. Comput. Appl. Math. 124, 209–228 (2000)
Wang, X., Zhou, G.: A new filled function for unconstrained global optimization. Appl. Comput. Math. 174, 419–429 (2006)
Chinchuluun, A., Rentsen, E., Pardalos, P.M.: Global minimization algorithms for concave quadratic programming problems. Optimization 24, 627–639 (2005)
Toh, K.: Global optimization by monotonic transformation. Comput. Optim. Appl. 23, 77–99 (2002)
Zhu, W.X.: Dynamic globally concavized filled function method for continuous global optimization. J. Optim. Theory Appl. 139(3), 635–648 (2008)
Zhu, W.X., Ali, M.: Solving nonlinearly constrained global optimization problem via an auxiliary function method. J. Comput. Appl. Math. 230(2), 491–503 (2009)
Pardalos, P.M., eds, R.E.: Handbook of Global Optimization. Volume 2: Heuristic Approaches. Kluwer Academic, New York (2002)
Chelouah, R., Siarry, P.: Tabu search applied to global optimization. Eur. J. Oper. Res. 123, 256–270 (2000)
Bunnag, D.: Stochastic algorithms for global optimization. Ph.D. Thesis, The University of Alabama (2004)
Wang, T.: Global optimization for constrained nonlinear programming. Ph.D. Thesis, University of Illinois at Urbana-Champaign (2001)
Chew, S., Zheng, Q.: Integral Global Optimization (Theory, Implementation and Applications). Springer, Berlin (1988)
Tian, W., Wu, D., et al.: Modified integral-level set method for the constrained global optimization. Appl. Math. Mech. 25, 202–210 (2004)
Wu, D., Tian, W., et al.: An algorithm of modified integral-level set for solving global optimization. Acta Math. Appl. Sin. 24, 100–110 (2001)
Phu, H.X., Hoffmann, A.: Essential supremum and supremum of summable function. Numer. Funct. Anal. Optim. 17, 167–180 (1996)
Hichert, J.: Methoden zur Bestimmung des wesentlichen Supremums mit anwendung in der globalen Optimierung. Thesis TU Ilmenau, N231 (1999) (also published in Shaker Verlag, Aachen (2000). ISBN 3-8265-8206-3)
Hichert, J., Hoffmann, A., Phu, H.X.: Convergence speed of an integral method for computing the essential supremum. In: Bomze, I.M., Csendes, T., Horst, R., Pardalos, P.M. (eds.) Developments in global Optimization, pp. 153–170. Kluwer Academic, New York (1997)
Hichert, J., Hoffmann, A., Phu, H.X., Reinhardt, R.: A primal-dual integral method in global optimization. Discuss. Math., Differ. Incl. Control Optim. 20, 257–278 (2000)
Wu, D., Yu, W., Tian, W.: A level-value estimation method for solving global optimization. Appl. Math. Mech. 27, 1001–1010 (2006)
Peng, Z., Wu, D., Tian, W.: A level-value estimation method for solving constrained global optimization. Math. Numer. Sin. 29, 293–304 (2007)
Asmussen, S., Kroese, D.P., Rubinstein, R.Y.: Heavy tails, importance sampling and cross-entropy. Stoch. Models 21, 57–76 (2005)
Ross, S.M.: Simulation, 3rd edn. Posts and Telecom Press, Beijing (2006)
De Boer, P.T., Kroese, D.P., et al.: A tutorial on the cross-entropy method. Ann. Oper. Res. 134, 19–67 (2005)
Kroese, D.P., Porotsky, S., Rubinstein, R.Y.: The cross-entropy method for continuous multi-extremal optimization. Methodol. Comput. Appl. Probab. 8, 383–407 (2006)
Rubinstein, R.Y.: The cross-entropy method for combinatorial and continuous optimization. Methodol. Comput. Appl. Probab. 2, 127–190 (1999)
Dembo, R.S., Eisenstat, S.C., Steihaug, T.: Inexact Newton methods. SIAM J. Numer. Anal. 19, 400–408 (1982)
Nocedal, J., Wright, S.J.: Numerical Optimization. Springer, New York (1999)
Robert, C.P., Casella, G.: Monte Carlo Statistical Methods, 2nd edn. Springer, Berlin (2004)
Hu, J., Fu, M., Marcus, S.I.: Model reference adaptive search method for global optimization. Oper. Res. 55, 549–568 (2007)
Gong, G., Qian, M.: A Course in Applied Stochastic Process and Stochastic Models in Algorithms and Intelligent Computing. Tsinghua University Press, Beijing (2004)
Paditz, L.: On the analytical structure of the constant in the nonuniform version of the essen inequality. Statistics 20(3), 453–464 (1989)
Luenberger, D.G.: Linear and Nonlinear Programming, 2nd edn. Addison-Wesley, Reading (1984)
Floudas, C.A., Pardalos, P.M., Adjiman, C.S., et al.: Handbook of Test Problems for Local and Global Optimization. Kluwer Academic, New York (1999)
Mishra, S.K.: Some new test functions for global optimization and performance of repulsive particle swarm method. Available at SSRN: http://ssrn.com/abstract=926132
Molga, M., Smutnicki, C.: Test functions for optimization needs. www.zsd.ict.pwr.wroc.pl/files/docs/functions.pdf
Acknowledgements
The authors would like to thank Professor A. Hoffmann and Professor C.-S. Lee for their help in improving the quality of the initial manuscript. They are very grateful to Professor Q.-C. Zhao and the anonymous reviewers for their constructive comments which play a crucial role on the final version of this paper. This work was supported by Natural Science Foundation of China (61170308), the Natural Science Foundation of FuJian Province (2011J01008) and FuJian Province Eduction Department (JA11033).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Qianchuan Zhao.
Rights and permissions
About this article
Cite this article
Peng, Z., Wu, D. & Zheng, Q. A Level-Value Estimation Method and Stochastic Implementation for Global Optimization. J Optim Theory Appl 156, 493–523 (2013). https://doi.org/10.1007/s10957-012-0151-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10957-012-0151-1