Quantifying uncertainties on excursion sets under a Gaussian random field prior
From MaRDI portal
Abstract: We focus on the problem of estimating and quantifying uncertainties on the excursion set of a function under a limited evaluation budget. We adopt a Bayesian approach where the objective function is assumed to be a realization of a Gaussian random field. In this setting, the posterior distribution on the objective function gives rise to a posterior distribution on excursion sets. Several approaches exist to summarize the distribution of such sets based on random closed set theory. While the recently proposed Vorob'ev approach exploits analytical formulae, further notions of variability require Monte Carlo estimators relying on Gaussian random field conditional simulations. In the present work we propose a method to choose Monte Carlo simulation points and obtain quasi-realizations of the conditional field at fine designs through affine predictors. The points are chosen optimally in the sense that they minimize the posterior expected distance in measure between the excursion set and its reconstruction. The proposed method reduces the computational costs due to Monte Carlo simulations and enables the computation of quasi-realizations on fine designs in large dimensions. We apply this reconstruction approach to obtain realizations of an excursion set on a fine grid which allow us to give a new measure of uncertainty based on the distance transform of the excursion set. Finally we present a safety engineering test case where the simulation method is employed to compute a Monte Carlo estimate of a contour line.
Recommendations
- Efficient Monte Carlo for high excursions of Gaussian random fields
- A SUR version of the Bichon criterion for excursion set estimation
- Sequential design of computer experiments for the estimation of a probability of failure
- Excursion and contour uncertainty regions for latent Gaussian models
- Fast sampling of parameterised Gaussian random fields
Cites work
- scientific article; zbMATH DE number 1692928 (Why is no real title available?)
- scientific article; zbMATH DE number 5722226 (Why is no real title available?)
- scientific article; zbMATH DE number 3913340 (Why is no real title available?)
- scientific article; zbMATH DE number 4086810 (Why is no real title available?)
- scientific article; zbMATH DE number 52492 (Why is no real title available?)
- A Limited Memory Algorithm for Bound Constrained Optimization
- Algorithm 659
- An informational approach to the global optimization of expensive-to-evaluate functions
- Confidence regions for means of random sets using oriented distance functions
- Convergence properties of the expected improvement algorithm with fixed mean and covariance functions
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Distance transforms of sampled functions
- Efficient global optimization of expensive black-box functions
- Excursion and contour uncertainty regions for latent Gaussian models
- Expectations of random sets and their boundaries using oriented distance functions
- Finding near-optimal Bayesian experimental designs via genetic algorithms
- Gaussian processes for machine learning.
- Geostatistics. Modeling spatial uncertainty.
- Interpolation of spatial data. Some theory for kriging
- Large Sample Properties of Simulations Using Latin Hypercube Sampling
- On excursion sets, tube formulas and maxima of random fields.
- Quantifying uncertainty on Pareto fronts with Gaussian process conditional simulations
- Sequential design of computer experiments for the estimation of a probability of failure
- Sequential methods for design-adaptive estimation of discontinuities in regression curves and surfaces
- Set Reconstruction by Voronoi Cells
- Spatio-temporal exceedance locations and confidence regions
- Theory of Random Sets
- \textit{KrigInv}: an efficient and user-friendly implementation of batch-sequential inversion strategies based on kriging
Cited in
(17)- Extrapolation of stationary random fields via level sets
- Excursion and contour uncertainty regions for latent Gaussian models
- Locally induced Gaussian processes for large-scale simulation experiments
- Uncertainty Quantification and Experimental Design for Large-Scale Linear Inverse Problems under Gaussian Process Priors
- Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling
- Generalized probabilistic bisection for stochastic root finding
- pGPx
- Forecast evaluation of quantiles, prediction intervals, and other set-valued functionals
- Profile Extrema for Visualizing and Quantifying Uncertainties on Excursion Regions: Application to Coastal Flooding
- Using an autonomous underwater vehicle with onboard stochastic advection-diffusion models to map excursion sets of environmental variables
- Adaptive Design of Experiments for Conservative Estimation of Excursion Sets
- A multifidelity quantile-based approach for confidence sets of random excursion sets with application to ice-sheet dynamics
- Evaluating Gaussian process metamodels and sequential designs for noisy level set estimation
- A SUR version of the Bichon criterion for excursion set estimation
- Probabilistic bisection with spatial metamodels
- Non-parametric measure approximations for constrained multi-objective optimisation under uncertainty
- Gauss von Mises Distribution for Improved Uncertainty Realism in Space Situational Awareness
Describes a project that uses
Uses Software
This page was built for publication: Quantifying uncertainties on excursion sets under a Gaussian random field prior
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q97314)
