Adaptive design and analysis of supercomputer experiments

TL;DR

This paper introduces an adaptive Bayesian treed Gaussian process method for efficiently exploring and modeling complex response surfaces in supercomputer experiments, optimizing the use of computational resources.

Contribution

It develops a novel adaptive sequential design framework combining Bayesian treed GPs with active learning strategies for supercomputing environments.

Findings

Effective exploration of response surfaces demonstrated in fluid dynamics simulations

Adaptive design reduces computational costs compared to fixed designs

Method provides explicit uncertainty quantification

Abstract

Computer experiments are often performed to allow modeling of a response surface of a physical experiment that can be too costly or difficult to run except using a simulator. Running the experiment over a dense grid can be prohibitively expensive, yet running over a sparse design chosen in advance can result in obtaining insufficient information in parts of the space, particularly when the surface calls for a nonstationary model. We propose an approach that automatically explores the space while simultaneously fitting the response surface, using predictive uncertainty to guide subsequent experimental runs. The newly developed Bayesian treed Gaussian process is used as the surrogate model, and a fully Bayesian approach allows explicit measures of uncertainty. We develop an adaptive sequential design framework to cope with an asynchronous, random, agent--based supercomputing environment, by using a hybrid approach that melds optimal strategies from the statistics literature with flexible strategies from the active learning literature. The merits of this approach are borne out in several examples, including the motivating computational fluid dynamics simulation of a rocket booster.