Emulation of multivariate simulators using thin-plate splines with application to atmospheric dispersion
From MaRDI portal
Abstract: It is often desirable to build a statistical emulator of a complex computer simulator in order to perform analysis which would otherwise be computationally infeasible. We propose methodology to model multivariate output from a computer simulator taking into account output structure in the responses. The utility of this approach is demonstrated by applying it to a chemical and biological hazard prediction model. Predicting the hazard area that results from an accidental or deliberate chemical or biological release is imperative in civil and military planning and also in emergency response. The hazard area resulting from such a release is highly structured in space and we therefore propose the use of a thin-plate spline to capture the spatial structure and fit a Gaussian process emulator to the coefficients of the resultant basis functions. We compare and contrast four different techniques for emulating multivariate output: dimension-reduction using (i) a fully Bayesian approach with a principal component basis, (ii) a fully Bayesian approach with a thin-plate spline basis, assuming that the basis coefficients are independent, and (iii) a "plug-in" Bayesian approach with a thin-plate spline basis and a separable covariance structure; and (iv) a functional data modeling approach using a tensor-product (separable) Gaussian process. We develop methodology for the two thin-plate spline emulators and demonstrate that these emulators significantly outperform the principal component emulator. Further, the separable thin-plate spline emulator, which accounts for the dependence between basis coefficients, provides substantially more realistic quantification of uncertainty, and is also computationally more tractable, allowing fast emulation. For high resolution output data, it also offers substantial predictive and computational advantages over the tensor-product Gaussian process emulator.
Recommendations
- Inferring atmospheric release characteristics in a large computer experiment using Bayesian adaptive splines
- Parallel partial Gaussian process emulation for computer models with massive output
- Sensitivity analysis and emulation for functional data using Bayesian adaptive splines
- Manifold learning for the emulation of spatial fields from computational models
- Using emulators to estimate uncertainty in complex models
Cites work
- Bayesian calibration of computer models. (With discussion)
- Bayesian emulation of complex multi-output and dynamic computer models
- Bayesian Updating of Atmospheric Dispersion After a Nuclear Accident
- Cases for the nugget in modeling computer experiments
- Computer Model Calibration Using High-Dimensional Output
- Computer model validation with functional output
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Exploratory designs for computational experiments
- Gaussian processes for machine learning.
- scientific article; zbMATH DE number 1077338 (Why is no real title available?)
- scientific article; zbMATH DE number 1865509 (Why is no real title available?)
- scientific article; zbMATH DE number 2231192 (Why is no real title available?)
- Multi-fidelity optimization via surrogate modelling
- Optimal design for additive partially nonlinear models
- Principal component analysis.
- Spline-Based Emulators for Radiative Shock Experiments With Measurement Error
- The design and analysis of computer experiments.
- Thin Plate Regression Splines
Cited in
(6)- Sensitivity analysis and emulation for functional data using Bayesian adaptive splines
- Inferring atmospheric release characteristics in a large computer experiment using Bayesian adaptive splines
- Efficient emulation of computer models utilising multiple known boundaries of differing dimension
- Gaussian Process Modeling of a Functional Output with Information from Boundary and Initial Conditions and Analytical Approximations
- Gaussian Process Emulation for High-Dimensional Coupled Systems
- Tensor-variate Gaussian process regression for efficient emulation of complex systems: comparing regressor and covariance structures in outer product and parallel partial emulators
This page was built for publication: Emulation of multivariate simulators using thin-plate splines with application to atmospheric dispersion
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3179332)
