Research

Large-Scale Fully-Sequential Space-lling Algorithms for Computer Experiments

Joint work with Dr. Apley.

Fully-sequential (i.e., with design points added one-at-a-time) space-filling designs are useful for global surrogate modeling of expensive computer experiments when the number of design points required to achieve a suitable accuracy is unknown in advance. We develop and investigate a fully-sequential space-filling (FSSF) design algorithm that is conceptually simple and computationally efficient and that achieves much better space-filling properties than alternative methods such as Sobol sequences and more complex block-sequential methods based on sliced or nested Latin hypercube designs (LHDs). Remarkably, at each design size in the sequence, our FSSF algorithm even achieves much better space filling properties than a one-shot LHD optimized for that specific size. The algorithm also scales well to high-dimensional design spaces and very large design sizes. We provided a R package for the FSSF algorithm written in C++.

Diversity Subsampling: Uniform Samples from Nonuniform Data Sets

Joint work with Dr. Apley and Dr. Mehrotra.

To be updated.