English

Practical heteroskedastic Gaussian process modeling for large simulation experiments

Methodology 2019-01-18 v2 Computation

Abstract

We present a unified view of likelihood based Gaussian progress regression for simulation experiments exhibiting input-dependent noise. Replication plays an important role in that context, however previous methods leveraging replicates have either ignored the computational savings that come from such design, or have short-cut full likelihood-based inference to remain tractable. Starting with homoskedastic processes, we show how multiple applications of a well-known Woodbury identity facilitate inference for all parameters under the likelihood (without approximation), bypassing the typical full-data sized calculations. We then borrow a latent-variable idea from machine learning to address heteroskedasticity, adapting it to work within the same thrifty inferential framework, thereby simultaneously leveraging the computational and statistical efficiency of designs with replication. The result is an inferential scheme that can be characterized as single objective function, complete with closed form derivatives, for rapid library-based optimization. Illustrations are provided, including real-world simulation experiments from manufacturing and the management of epidemics.

Keywords

Cite

@article{arxiv.1611.05902,
  title  = {Practical heteroskedastic Gaussian process modeling for large simulation experiments},
  author = {Mickael Binois and Robert B. Gramacy and Michael Ludkovski},
  journal= {arXiv preprint arXiv:1611.05902},
  year   = {2019}
}

Comments

33 pages, 7 figures

R2 v1 2026-06-22T16:56:26.453Z