Efficient emulators of computer experiments using compactly supported correlation functions, with an application to cosmology
From MaRDI portal
Publication:765992
DOI10.1214/11-AOAS489zbMATH Open1234.62166arXiv1107.0749OpenAlexW3098133464MaRDI QIDQ765992FDOQ765992
Authors: Cari G. Kaufman, Salman Habib, Katrin Heitmann, Joshua A. Frieman, Derek Bingham 
Publication date: 22 March 2012
Published in: The Annals of Applied Statistics (Search for Journal in Brave)
Abstract: Statistical emulators of computer simulators have proven to be useful in a variety of applications. The widely adopted model for emulator building, using a Gaussian process model with strictly positive correlation function, is computationally intractable when the number of simulator evaluations is large. We propose a new model that uses a combination of low-order regression terms and compactly supported correlation functions to recreate the desired predictive behavior of the emulator at a fraction of the computational cost. Following the usual approach of taking the correlation to be a product of correlations in each input dimension, we show how to impose restrictions on the ranges of the correlations, giving sparsity, while also allowing the ranges to trade off against one another, thereby giving good predictive performance. We illustrate the method using data from a computer simulator of photometric redshift with 20,000 simulator evaluations and 80,000 predictions.
Full work available at URL: https://arxiv.org/abs/1107.0749
Recommendations
- A fast and calibrated computer model emulator: an empirical Bayes approach
- Flexible Correlation Structure for Accurate Prediction and Uncertainty Quantification in Bayesian Gaussian Process Emulation of a Computer Model
- Computer emulation with nonstationary Gaussian processes
- Compactly supported correlation functions
- Gaussian process emulators for computer experiments with inequality constraints
Bayesian inference (62F15) Inference from stochastic processes (62M99) Measures of association (correlation, canonical correlation, etc.) (62H20) Applications of statistics (62P99) Astrophysical cosmology (85A40)
Cites Work
- A modeling approach for large spatial datasets
- Quasi-regression
- The design and analysis of computer experiments.
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Compactly supported correlation functions
- Spatial designs and properties of spatial correlation: effects on covariance estimation
- Bayesian calibration of computer models. (With discussion)
- Objective Bayesian Analysis of Spatially Correlated Data
- Approximating Likelihoods for Large Spatial Data Sets
- Covariance tapering for likelihood-based estimation in large spatial data sets
- Default priors for Gaussioan processes
- A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code
- Title not available (Why is that?)
- Probabilistic Sensitivity Analysis of Complex Models: A Bayesian Approach
- Orthogonal Array-Based Latin Hypercubes
- Measurements of Ω and Λ from 42 High‐Redshift Supernovae
- On Abel-Poisson type and Riesz means
- Title not available (Why is that?)
- Criteria of Pólya type for radial positive definite functions
- Kriging with large data sets using sparse matrix techniques
- Bayesian prediction with adaptive ridge estimators
Cited In (33)
- Analysis methods for computer experiments: how to assess and what counts?
- Estimating percentiles in computer experiments: a comparison of sequential-adaptive designs and fixed designs
- Functional emulation of high resolution tsunami modelling over cascadia
- Calibrating a large computer experiment simulating radiative shock hydrodynamics
- Emulating satellite drag from large simulation experiments
- Uncertainty quantification for computer models with spatial output using calibration-optimal bases
- Locally induced Gaussian processes for large-scale simulation experiments
- Posterior belief assessment: extracting meaningful subjective judgements from Bayesian analyses with complex statistical models
- Uncertainty quantification of a computer model for binary black hole formation
- A physics-based emulator for the simulation of geophysical mass flows
- EzGP: Easy-to-Interpret Gaussian Process Models for Computer Experiments with Both Quantitative and Qualitative Factors
- Hierarchical sparse Cholesky decomposition with applications to high-dimensional spatio-temporal filtering
- Separable spatio-temporal kriging for fast virtual sensing
- Multiresolution functional ANOVA for large-scale, many-input computer experiments
- Adaptive design for Gaussian process regression under censoring
- A Global-Local Approximation Framework for Large-Scale Gaussian Process Modeling
- Spline-Based Emulators for Radiative Shock Experiments With Measurement Error
- Adaptive Design and Analysis Via Partitioning Trees for Emulation of a Complex Computer Code
- Hierarchical shrinkage Gaussian processes: applications to computer code emulation and dynamical system recovery
- Scaled Vecchia Approximation for Fast Computer-Model Emulation
- Vecchia-approximated Deep Gaussian Processes for Computer Experiments
- Practical Heteroscedastic Gaussian Process Modeling for Large Simulation Experiments
- On the Bayesian calibration of computer model mixtures through experimental data, and the design of predictive models
- A comparison of statistical emulation methodologies for multi‐wave calibration of environmental models
- Gaussian Process Subspace Prediction for Model Reduction
- A statistical method for emulation of computer models with invariance-preserving properties, with application to structural energy prediction
- Inferring atmospheric release characteristics in a large computer experiment using Bayesian adaptive splines
- Gaussian Process Prediction using Design-Based Subsampling
- Gaussian processes for computer experiments
- Efficient emulation of computer models utilising multiple known boundaries of differing dimension
- Locating infinite discontinuities in computer experiments
- Estimating sensitivity indices based on Gaussian process metamodels with compactly supported correlation functions
- Emulation of dynamic simulators with application to hydrology
Uses Software
This page was built for publication: Efficient emulators of computer experiments using compactly supported correlation functions, with an application to cosmology
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q765992)
