Skip to main content
Log in

A method of trust region type for minimizing noisy functions

Ein auf Vertrauensbereichen basierendes Verfahren zur Minimierung verrauschter Funktionen

  • Published:
Computing Aims and scope Submit manuscript

Abstract

The optimization of noisy functions in a few variables only is a common problem occurring in various applications, for instance in finding the optimal choice of a few control parameters in chemical experiments. The traditional tool for the treatment of such problems is the method of Nelder-Mead (NM). In this paper an alternative method based on a trust region approach (TR) is offered and compared to Nelder-Mead. On the standard collection of test functions for unconstrained optimization by Moré et al. [9], TR performs substantially more robust than NM. If performance is measured by the number of function evaluations, TR is on the average twice as fast as NM.

Zusammenfassung

Die Optimierung verrauschter Funktionen in einer geringen Zahl von Variablen ist ein häufig in Anwendungen vorkommendes Problem, z.B. um die optimale Wahl einiger Kontrollparameter in chemischen Experimenten zu finden. Das traditionelle Werkzeug zur Behandlung solcher Probleme ist das Verfahren von Nelder-Mead (NM). In dieser Arbeit wird eine auf Vertrauensbereichen beruhende Alternative (TR) vorgestellt und mit Nelder-Mead verglichen. Auf Moré et al.’s bekannter Sammlung von Testfunktionen für Optimierung ohne Nebenbedingungen arbeitet TR wesentlich robuster als NM. Mißt man die Qualität an der Zahl der Funktionsauswertungen, so ist TR im Mittel doppelt so schnell wie NM.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Brent, R. P.: Algorithms for minimization without derivatives. Englewood Cliffs: Prentice-Hall, 1973.

    MATH  Google Scholar 

  2. Dixon, L. C. W.: ACSIM — an accelerated constrained simplex technique. Comput. Aided Des.5, 22–32 (1973).

    Article  Google Scholar 

  3. Elster, C., Neumaier, A.: A grid algorithm for bound constrained optimization of noisy functions. IMA J. Numer. Anal.15, 585–608 (1995).

    Article  MATH  Google Scholar 

  4. Fletcher, R.: Practical methods of optimization, New York: Wiley, 1987.

    MATH  Google Scholar 

  5. Gill, E., Murray, W., Wright, M.H.: Practical optimization. London: Academic Press, 1981.

    MATH  Google Scholar 

  6. Glad, T., Goldstein, A.: Optimization of functions whose values are subject to small errors. BIT17, 160–169, (1977).

    Article  MATH  Google Scholar 

  7. Karidis, J. P., Turns, S. R.: Efficient optimization of computationally expensive objective functions. IBM Research Report RC10698 (#47972): Yorktown Heights, NY (1984).

  8. Moré, J. J.: Recent developments in algorithms and software for trust region methods. In: Mathematical programming: the state of the art (Bachem, A. et al., eds.), pp. 256–287. Berlin: Springer, 1983.

    Google Scholar 

  9. Moré, J. J., Garbow, B. S., Hillstrom, K. E.: Testing unconstrained optimization software. ACM Trans. Math. Software7, 17–41 (1981).

    Article  MATH  Google Scholar 

  10. Moré, J. J., Sorensen, D. C.: Computing a trust region step. SIAM J. Sci. Stat. Comput4, 553–572 (1983).

    Article  MATH  Google Scholar 

  11. Numerical Algorithm Group (NAG), Fortran Library Mark 14. Oxford, 1990.

  12. Nelder, J. A., Mead, R.: A simplex method for function minimization. Comput. J.7, 308 (1965).

    Google Scholar 

  13. Nocedal, J.: Theory of algorithms for unconstrained optimization. In: Acta numerica 1992 (Iserles, A., ed.), pp. 199–242. Cambridge: Cambridge University Press, 1992.

    Google Scholar 

  14. Powell, M. J. D.: A review of algorithms for nonlinear equations and unconstrained optimization. In: ICIAM 1987 Proceedings (McKenna, J., Temam, R., eds.), pp. 220–232. Philadelphia: SIAM, 1988.

    Google Scholar 

  15. Torczon, V.: On the convergence of the multidimensional search algorithm. SIAM J. Optim.1, 123–145 (1991).

    Article  MATH  Google Scholar 

  16. Winfield, D.: Function minimization by interpolation in a data table. J. Inst. Math. Appl.12, 339–348 (1973).

    Article  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Elster, C., Neumaier, A. A method of trust region type for minimizing noisy functions. Computing 58, 31–46 (1997). https://doi.org/10.1007/BF02684470

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02684470

AMS Subject Classification

Key words

Navigation