gaussian_process

Gaussian Process surrogate model

Specification

  • Alias: kriging

  • Arguments: None

  • Default: Surfpack Gaussian process

Child Keywords:

Required/Optional

Description of Group

Dakota Keyword

Dakota Keyword Description

Required (Choose One)

GP Implementation

surfpack

Use the Surfpack version of Gaussian Process surrogates

dakota

Select the built in Gaussian Process surrogate

Optional

build_samples

Number of initial model evaluations used in build phase

Optional

posterior_adaptive

Adapt emulator model to increase accuracy in high posterior probability regions

Optional

import_build_points_file

File containing points you wish to use to build a surrogate

Description

Use the Gaussian process (GP) surrogate from Surfpack, which is specified using the model-surrogate-global-gaussian_process-surfpack keyword.

An alternate version of GP surrogates was available in prior versions of Dakota. For now, both versions are supported but the ``dakota`` version is deprecated and intended to be removed in a future release.

Known Issue: When using discrete variables, there have been sometimes significant differences in surrogate behavior observed across computing platforms in some cases. The cause has not yet been fully diagnosed and is currently under investigation. In addition, guidance on appropriate construction and use of surrogates with discrete variables is under development. In the meantime, users should therefore be aware that there is a risk of inaccurate results when using surrogates with discrete variables.