This paper uses a sequentialized experimental design to select simulation input com- binations for global optimization, based on Kriging (also called Gaussian process or spatial correlation modeling); this Kriging is used to analyze the input/output data of the simulation model (computer code). This paper adapts the classic "ex- pected improvement" (EI) in "efficient global optimization" (EGO) through the introduction of an unbiased estimator of the Kriging predictor variance; this estima- tor uses parametric bootstrapping. Classic EI and bootstrapped EI are compared through four popular test functions, including the six-hump camel-back and two Hartmann functions. These empirical results demonstrate that in some applications bootstrapped EI finds the global optimum faster than classic EI does; in general, however, the classic EI may be considered to be a robust global optimizer.
|Place of Publication||Tilburg|
|Number of pages||17|
|Publication status||Published - 2010|
|Name||CentER Discussion Paper|