Abstract
For the optimization under uncertainty problem, there has been recent interest in coupling trust-region methods with surrogate surfaces or function approximations. There are many theoretical and statistical issues that must be carefully considered in following such an approach. Herein, the Nadaraya-Watson estimator is used for the smooth function approximation, and the effects of observation noise and random sampling on estimator error are examined. For the fundamental optimization problem where the exact function is quadratic, analytical results are derived for the mean-square error of the difference and gradient of the function. It is also shown how these statistics are related to the trust-region method, how the analytical results can be used to determine the bandwidth of the kernel of the estimator, and how third-order terms can affect the error statistics.
Similar content being viewed by others
References
Alexandrov N, JED Jr, Lewis RM, Torczon V (1997) A trust region framework for managing the use of approximation models in optimization. ICASE Report 97-50, NASA
Carter RG (1991) On the global convergence of trust region algorithms using inexact gradient information. J Numer Anal 28(1):251–265
Carter RG (1993) Numerical experience with a class of algorithms for nonlinear optimization using inexact fuction and gradient information. J Sci Comput 14:368–388
Collomb G (1985) Nonparametric regression—an up-to-date bibliography. Statistics 2:309–324
Conn AR, Scheinberg K, Toint P (1997) On the convergence of derivative-free methods for unconstrained opimization. In: Buhmann MD, Iserles A (eds) Approximation theory and optimization: tributes to M.J.D. Powell. Cambridge University Press, pp 83–108
Eldred M, Giunta A, Wojtkiewicz JSF, Trucano T (2002) Formulations for surrogate-based optimization under uncertainty. In: Proceedings of the 9th AIAA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, number paper AIAA-2002-5585, Atlanta, GA
Gasser T, Maller H (1984) Estimating regression functions and their derivatives by the kernel method. Scand J Stat 11:171–185
Giunta AA, Eldred MS (2000) Implementation of a trust region model management strategy in the DAKOTA optimization toolkit. In: Proceedings of the 8th AIAA/USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, Long Beach, CA
Hardle W (1990) Applied nonparametric regression, econometric society monographs No. 19. Cambridge University Press, London
Hardle W, Marron JS (1985) Optimal bandwidth selection in nonparametric regression function estimation. Ann Stat 13:1465–1481
More JJ (1983) Recent developments in algorithms and software for trust region methods. In: Bachem A, Grotschel M, Korte B (eds) Mathematical Programming, the State of the Art. Springer-Verlag, New York, pp 258–287
Nadaraya E (1964) On non-parametric estimates of density functions and regression curves. Theory Probab Appl 10:186–190
Rodriguez JF, Renaud JE, Watson LT (1997) Trust region augmented Lagrangian methods for sequential response surface approximation and optimization. In: Proceedings of the ASME Design Engineering Technical Conference
Stone C (1982) Optimal rates of convergence for nonparametric regression. Ann Stat 10:1040–1053
Trosset M, Alexandrov N, Watson L (2003) New methods for robust design using computer simulation. In: Proceedings of the Section on Physical and Engineering Sciences. American Statistical Association
Venter G, Haftka RT, Starnes JH (1998) Construction of response surface approximations for design optimization. AIAA Journal 36(12):2242–2249
Vieu P (1991) Nonparametric regression: Optimal local bandwidth choice. J R Stat Soc B, 453–464
Wand M (1992) Error analysis for general multivariate kernel estimatiors. Nonparametr Stat 2:1–15
Watson GS (1964) Smooth regression analysis. Sankhya 26:359–372
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wan, Z., Igusa, T. Statistics of Nadaraya-Watson estimator errors in surrogate-based optimization. Optim Eng 7, 385–397 (2006). https://doi.org/10.1007/s11081-006-9980-9
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/s11081-006-9980-9