Known Boundary Emulation of Complex Computer Models

DOI10.1137/18M1164457zbMATH Open1430.62238arXiv1801.03184OpenAlexW2963971501WikidataQ127519630 ScholiaQ127519630MaRDI QIDQ5237178FDOQ5237178

Authors: Samuel E. Jackson, Ian Vernon, Jonathan A. Cumming 

Publication date: 17 October 2019

Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Abstract: Computer models are now widely used across a range of scientific disciplines to describe various complex physical systems, however to perform full uncertainty quantification we often need to employ emulators. An emulator is a fast statistical construct that mimics the complex computer model, and greatly aids the vastly more computationally intensive uncertainty quantification calculations that a serious scientific analysis often requires. In some cases, the complex model can be solved far more efficiently for certain parameter settings, leading to boundaries or hyperplanes in the input parameter space where the model is essentially known. We show that for a large class of Gaussian process style emulators, multiple boundaries can be formally incorporated into the emulation process, by Bayesian updating of the emulators with respect to the boundaries, for trivial computational cost. The resulting updated emulator equations are given analytically. This leads to emulators that possess increased accuracy across large portions of the input parameter space. We also describe how a user can incorporate such boundaries within standard black box GP emulation packages that are currently available, without altering the core code. Appropriate designs of model runs in the presence of known boundaries are then analysed, with two kinds of general purpose designs proposed. We then apply the improved emulation and design methodology to an important systems biology model of hormonal crosstalk in Arabidopsis Thaliana.

Full work available at URL: https://arxiv.org/abs/1801.03184

Recommendations

• Efficient modelling techniques for complicated boundary conditions applied to structured grids
• A Computational Framework for Boundary-Value Problem Based Simulations
• Using emulators to estimate uncertainty in complex models
• scientific article; zbMATH DE number 4030918
• Accurate emulators for large-scale computer experiments

zbMATH Keywords

design of experimentssystems biologyboundary conditionsGaussian process emulationBayes linear emulation

Mathematics Subject Classification ID

Bayesian inference (62F15) Applications of statistics to biology and medical sciences; meta analysis (62P10) Systems biology, networks (92C42)

Cites Work

• The design and analysis of computer experiments.
• Design and analysis of computer experiments. With comments and a rejoinder by the authors
• Bayesian calibration of computer models. (With discussion)
• Computer Model Calibration Using High-Dimensional Output
• Title not available (Why is that?)
• Approximate Bayesian computation and simulation-based inference for complex stochastic epidemic models
• Galaxy formation: Bayesian history matching for the observable Universe
• Galaxy formation: a Bayesian uncertainty analysis
• Rejoinder: ``Galaxy formation: a Bayesian uncertainty analysis
• Bayes Linear Statistics
• Title not available (Why is that?)
• Efficient history matching of a high dimensional individual-based HIV transmission model
• Gaussian process modeling with boundary information

Cited In (5)

• Multi-fidelity Gaussian process modeling with boundary information
• Coupling computer models through linking their statistical emulators
• Gaussian Process Regression on Nested Spaces
• Using emulators to estimate uncertainty in complex models
• Efficient emulation of computer models utilising multiple known boundaries of differing dimension

Uses Software

• GPfit
• GPy
• GitHub

This page was built for publication: Known Boundary Emulation of Complex Computer Models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5237178)