Advertisement

Optimization and Engineering

, Volume 1, Issue 4, pp 373–397 | Cite as

Global Optimization of Costly Nonconvex Functions Using Radial Basis Functions

  • Mattias Björkman
  • Kenneth Holmström
Article

Abstract

The paper considers global optimization of costly objective functions, i.e. the problem of finding the global minimum when there are several local minima and each function value takes considerable CPU time to compute. Such problems often arise in industrial and financial applications, where a function value could be a result of a time-consuming computer simulation or optimization. Derivatives are most often hard to obtain, and the algorithms presented make no use of such information.

Several algorithms to handle the global optimization problem are described, but the emphasis is on a new method by Gutmann and Powell, A radial basis function method for global optimization. This method is a response surface method, similar to the Efficient Global Optimization (EGO) method of Jones. Our Matlab implementation of the Radial Basis Function (RBF) method is described in detail and we analyze its efficiency on the standard test problem set of Dixon-Szegö, as well as its applicability on a real life industrial problem from train design optimization. The results show that our implementation of the RBF algorithm is very efficient on the standard test problems compared to other known solvers, but even more interesting, it performs extremely well on the train design optimization problem.

global optimization nonconvex optimization software black box 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. M. Björkman, “Global optimization of costly nonconvex functions using radial basis functions,” Master thesis, Department of Mathematics and Physics, Mälardalen University, Sweden, 2000.Google Scholar
  2. M. Björkman and K. Holmström, “Global optimization using the DIRECT algorithm in Matlab,” Advanced Modeling and Optimization vol. 1, no. 2, pp. 17-37, 1999.Google Scholar
  3. L. C. W. Dixon and G. P. Szegö, “The global optimisation problem: An introduction,” in Toward Global Optimization, L. Dixon and G. Szego, eds., North Holland Publishing Company: New York, 1978, pp. 1-15.Google Scholar
  4. R. Enblom, “Procedure for mechanical train set optimization using a two-level collaborative optimization approach,” Master thesis, Department of Mathematics and Physics, Mälardalen University, Sweden, 1998.Google Scholar
  5. J. Gablonsky, “An implementation of the DIRECT algorithm,” Technical Report, North Carolina State University, Department for Mathematics, 1998.Google Scholar
  6. H.-M. Gutmann, “A radial basis for function method for global optimization,” Talk at IFIP TC7 Conference, Cambridge, 1999.Google Scholar
  7. H.-M. Gutmann, “A radial basis function method for global optimization,” Journal of Global Optimization vol. 19, no. 3, pp. 201-227, 2001.Google Scholar
  8. H.-M. Gutmann, “On the semi-norm of radial basis function interpolants,” Technical Report DAMTP 2000/NA04, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, England, 2000.Google Scholar
  9. T. Hellström and K. Holmström, “Parameter tuning in trading algorithms using ASTA,” in Computational Finance 1999, Y. S. Abu-Mostafa, B. LeBaron, A. W. Lo, and A. S. Weigand, eds., MIT Press: Cambridge, MA, 1999.Google Scholar
  10. K. Holmström, “New optimization algorithms and software,” Theory of Stochastic Processes vol. 5(21), no. 1-2, pp. 55-63, 1999a.Google Scholar
  11. K. Holmström, “The TOMLAB optimization environment in Matlab,” Advanced Modeling and Optimization vol. 1, no. 1, pp. 47-69, 1999b.Google Scholar
  12. K. Holmström, “The TOMLAB v2.0 optimization environment,” in Sixth Meeting of the Nordic Section of the Mathematical Programming Society, Västerås, E. Dotzauer, M. Björkman, and K. Holmström eds., 1999c.Google Scholar
  13. K. Holmström, “TOMLAB v3.0 user's guide,” Technical Report IMa-TOM-2001-01, Department of Mathematics and Physics, Mälardalen University, Sweden, 2001.Google Scholar
  14. K. Holmström and M. Björkman, “The TOMLAB NLPLIB toolbox for nonlinear programming,” Advanced Modeling and Optimization vol. 1, no. 1, pp. 70-86, 1999.Google Scholar
  15. R. Horst and P. M. Pardalos, Handbook of Global Optimization, Kluwer Academic Publishers: Dordrecht, 1995.Google Scholar
  16. W. Huyer and A. Neumaier, “Global optimization by multilevel coordinate search,” Journal of Global Optimization vol. 14, pp. 331-355.Google Scholar
  17. D. R. Jones, “DIRECT,” Encyclopedia of Optimization, to be published.Google Scholar
  18. D. R. Jones, C. D. Perttunen, and B. E. Stuckman, “Lipschitzian optimization without the Lipschitz constant,” Journal of Optimization Theory and Applications vol. 79, no. 1, pp. 157-181, 1993.Google Scholar
  19. D. R. Jones, M. Schonlau, and W. J. Welch, “Efficient global optimization of expensive black-box functions,” Journal of Global Optimization vol. 13, pp. 455-492, 1998.Google Scholar
  20. S. A. Nelson II and P. Y. Papalambros, “A modification to Jones' global optimization algorithm for fast local convergence,” Technical Report AIAA-98-4751, American Institute of Aeronautics and Astronautics, 1998.Google Scholar
  21. I. Persson, “The GENSYS.9903 User's Manual,” 1999.Google Scholar
  22. M. J. D. Powell, “The theory of radial basis function approximation in 1990,” in Advances in Numerical Analysis, pp. 105-210, Volume 2: Wavelets, Subdivision Algorithms and Radial Basis Functions, N. Light, ed., Oxford University Press: 1992, pp. 105-210.Google Scholar
  23. M. J. D. Powell, “A review of algorithms for thin plate spline interpolation in two dimensions,” in Advanced Topics in Multivariate Approximation, F. Fontanella, K. Jetter, and P. J. Laurent eds., World Scientific Publishing: River Edge, NJ., 1996, pp. 303-322.Google Scholar
  24. M. J. D. Powell, “Recent research at Cambridge on radial basis functions,” in New Developments in Approximation Theory, M. D. Buhmann, M. Felten, D. Mache, and M. W. Müller eds., Birkhäuser: Basel, 1999, pp. 215-232.Google Scholar

Copyright information

© Kluwer Academic Publishers 2000

Authors and Affiliations

  • Mattias Björkman
    • 1
  • Kenneth Holmström
    • 2
  1. 1.Center for Mathematical Modeling, Department of Mathematics and PhysicsMälardalen UniversityVästeråsSweden
  2. 2.Center for Mathematical Modeling, Department of Mathematics and PhysicsMälardalen UniversityVästeråsSweden

Personalised recommendations