Доступ предоставлен для: Guest
Портал Begell Электронная Бибилиотека e-Книги Журналы Справочники и Сборники статей Коллекции
International Journal for Uncertainty Quantification
Импакт фактор: 3.259 5-летний Импакт фактор: 2.547 SJR: 0.417 SNIP: 0.8 CiteScore™: 1.52

ISSN Печать: 2152-5080
ISSN Онлайн: 2152-5099

Свободный доступ

International Journal for Uncertainty Quantification

DOI: 10.1615/Int.J.UncertaintyQuantification.2013006809
pages 205-223

IMPROVEMENTS TO GRADIENT-ENHANCED KRIGING USING A BAYESIAN INTERPRETATION

Jouke H.S. de Baar
TU Delft, Kluyverweg 1 (10.18), 2629 HS Delft, The Netherlands
Richard P. Dwight
Aerodynamics Group, Faculty of Aerospace, TU Delft, P.O. Box 5058, 2600GB Delft, The Netherlands
Hester Bijl
TU Delft, Kluyverweg 1 (10.18), 2629 HS Delft, The Netherlands

Краткое описание

Cokriging is a flexible tool for constructing surrogate models on the outputs of computer models. It can readily incorporate gradient information, in which form it is named gradient-enhanced Kriging (GEK), and promises accurate surrogate models in >10 dimensions with a moderate number of sample locations for sufficiently smooth responses. However, GEK suffers from several problems: poor robustness and ill-conditionedness of the surface. Furthermore it is unclear how to account for errors in gradients, which are typically larger than errors in values. In this work we derive GEK using Bayes' Theorem, which gives an useful interpretation of the method, allowing construction of a gradient-error contribution. The Bayesian interpretation suggests the "observation error" as a proxy for errors in the output of the computer model. From this point we derive analytic estimates of robustness of the method, which can easily be used to compute upper bounds on the correlation range and lower bounds on the observation error. We thus see that by including the observation error, treatment of errors and robustness go hand in hand. The resulting GEK method is applied to uncertainty quantification for two test problems.


Articles with similar content:

A GRADIENT-BASED SAMPLING APPROACH FOR DIMENSION REDUCTION OF PARTIAL DIFFERENTIAL EQUATIONS WITH STOCHASTIC COEFFICIENTS
International Journal for Uncertainty Quantification, Vol.5, 2015, issue 1
Miroslav Stoyanov, Clayton G. Webster
FAST AND FLEXIBLE UNCERTAINTY QUANTIFICATION THROUGH A DATA-DRIVEN SURROGATE MODEL
International Journal for Uncertainty Quantification, Vol.8, 2018, issue 2
Gerta Köster, Hans-Joachim Bungartz, Felix Dietrich, Tobias Neckel, Florian Künzner
BIVARIATE QUANTILE INTERPOLATION FOR ENSEMBLE DERIVED PROBABILITY DENSITY ESTIMATES
International Journal for Uncertainty Quantification, Vol.5, 2015, issue 2
Brad Eric Hollister, Alex Pang
AN OVERVIEW OF INVERSE MATERIAL IDENTIFICATION WITHIN THE FRAMEWORKS OF DETERMINISTIC AND STOCHASTIC PARAMETER ESTIMATION
International Journal for Uncertainty Quantification, Vol.3, 2013, issue 4
Miguel A. Aguilo, Laura P. Swiler, Angel Urbina
Reconstruction of the Model of Probabilistic Dependences by Statistical Data. Tools and Algorithm
Journal of Automation and Information Sciences, Vol.41, 2009, issue 12
Alexander S. Balabanov