Abstract:
|
In the past, computer experiments simulations have been performed with the use of a single computer. To take advantage of the having multiple computers, these simulations need to be run in batches with the batch size determined by the number of computers. Various methods to perform these simulations in batches are proposed that utilize a combination of expected improvement and space filling criteria in order to construct efficient experimental designs for sequentially-adaptive computer experiments problems. These methods begin the sampling process by selecting the site with the best expected improvement, and then utilize various methods that balance high expected improvement with good performance under some space-filling criterion. These methods are tested on the problems of global fit and optimization, and the results are compared and contrasted with each other, as well as with one-at-a-time methods and other batch methods in the literature, to evaluate the effectiveness of the proposed methods for sampling in batches.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.