Emulators for stochastic simulation codes

DOI10.1051/PROC/201448005MaRDI QIDQ5744916zbMATH OpenOpenAlexFDO

Authors Vincent Moutoussamy, Simon Nanty, Benoît Pauwels

Publication date 10 February 2016

Published in ESAIM: Proceedings and Surveys (Search for Journal in Brave)

Copyright license Creative Commons Attribution-NonCommercial 4.0 International

Full work available at URL https://arxiv.org/abs/1406.6348

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Factor analysis and principal components; correspondence analysis (62H25) Design of statistical experiments (62K99)

Abstract: Numerical simulation codes are very common tools to study complex phenomena, but they are often time-consuming and considered as black boxes. For some statistical studies (e.g. asset management, sensitivity analysis) or optimization problems (e.g. tuning of a molecular model), a high number of runs of such codes is needed. Therefore it is more convenient to build a fast-running approximation - or metamodel - of this code based on a design of experiments. The topic of this paper is the definition of metamodels for stochastic codes. Contrary to deterministic codes, stochastic codes can give different results when they are called several times with the same input. In this paper, two approaches are proposed to build a metamodel of the probability density function of a stochastic code output. The first one is based on kernel regression and the second one consists in decomposing the output density on a basis of well-chosen probability density functions, with a metamodel linking the coefficients and the input parameters. For the second approach, two types of decomposition are proposed, but no metamodel has been designed for the coefficients yet. This is a topic of future research. These methods are applied to two analytical models and three industrial cases.

Recommendations

• Gaussian process emulators for the stochastic finite element method
• Emulation of stochastic simulators using generalized lambda models
• Stochastic kriging for simulation metamodeling
• Gaussian process emulation of dynamic computer codes
• Sampling, metamodeling, and sensitivity analysis of numerical simulators with functional stochastic inputs

Cites work

• scientific article; zbMATH DE number 3772748 (Why is no real title available?)
• scientific article; zbMATH DE number 3222478 (Why is no real title available?)
• A Limited Memory Algorithm for Bound Constrained Optimization
• A numerically stable dual method for solving strictly convex quadratic programs
• Comparison of Some Reduced Representation Approximations
• Dimensionality reduction when data are density functions
• Gaussian process modeling with inequality constraints
• Optimal control of parameter-dependent convection-diffusion problems around rigid bodies
• Remarks on Some Nonparametric Estimates of a Density Function
• Stochastic kriging for simulation metamodeling
• Variable kernel density estimation

Cited in

(9)

• Gaussian process emulators for the stochastic finite element method
• Sparse polynomial chaos expansion for universal stochastic kriging
• Analyzing stochastic computer models: a review with opportunities
• Polynomial chaos surrogate construction for random fields with parametric uncertainty
• Population parametrization of costly black box models using iterations between SAEM algorithm and Kriging
• Emulation of stochastic simulators using generalized lambda models
• Global sensitivity analysis and Wasserstein spaces
• A spectral surrogate model for stochastic simulators computed from trajectory samples
• Accurate emulators for large-scale computer experiments

Describes a project that uses

Uses Software

• LBFGS-B

This page was built for publication: Emulators for stochastic simulation codes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5744916)