hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R
An increasing number of time-consuming simulators exhibit a complex noise structure that depends on the inputs. For conducting studies with limited budgets of evaluations, new surrogate methods are required in order to simultaneously model the mean and variance fields. To this end, we present the he...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Foundation for Open Access Statistics
2021-07-01
|
Series: | Journal of Statistical Software |
Subjects: | |
Online Access: | https://www.jstatsoft.org/index.php/jss/article/view/3655 |
_version_ | 1797813924297441280 |
---|---|
author | Mickaël Binois Robert B. Gramacy |
author_facet | Mickaël Binois Robert B. Gramacy |
author_sort | Mickaël Binois |
collection | DOAJ |
description | An increasing number of time-consuming simulators exhibit a complex noise structure that depends on the inputs. For conducting studies with limited budgets of evaluations, new surrogate methods are required in order to simultaneously model the mean and variance fields. To this end, we present the hetGP package, implementing many recent advances in Gaussian process modeling with input-dependent noise. First, we describe a simple, yet efficient, joint modeling framework that relies on replication for both speed and accuracy. Then we tackle the issue of data acquisition leveraging replication and exploration in a sequential manner for various goals, such as for obtaining a globally accurate model, for optimization, or for contour finding. Reproducible illustrations are provided throughout. |
first_indexed | 2024-03-13T07:59:52Z |
format | Article |
id | doaj.art-29f17e8e7fe946399c4da3994eec01ad |
institution | Directory Open Access Journal |
issn | 1548-7660 |
language | English |
last_indexed | 2024-03-13T07:59:52Z |
publishDate | 2021-07-01 |
publisher | Foundation for Open Access Statistics |
record_format | Article |
series | Journal of Statistical Software |
spelling | doaj.art-29f17e8e7fe946399c4da3994eec01ad2023-06-01T18:41:06ZengFoundation for Open Access StatisticsJournal of Statistical Software1548-76602021-07-0198110.18637/jss.v098.i133496hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in RMickaël BinoisRobert B. GramacyAn increasing number of time-consuming simulators exhibit a complex noise structure that depends on the inputs. For conducting studies with limited budgets of evaluations, new surrogate methods are required in order to simultaneously model the mean and variance fields. To this end, we present the hetGP package, implementing many recent advances in Gaussian process modeling with input-dependent noise. First, we describe a simple, yet efficient, joint modeling framework that relies on replication for both speed and accuracy. Then we tackle the issue of data acquisition leveraging replication and exploration in a sequential manner for various goals, such as for obtaining a globally accurate model, for optimization, or for contour finding. Reproducible illustrations are provided throughout.https://www.jstatsoft.org/index.php/jss/article/view/3655input-dependent noiselevel-set estimationoptimizationreplicationstochastic kriging |
spellingShingle | Mickaël Binois Robert B. Gramacy hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R Journal of Statistical Software input-dependent noise level-set estimation optimization replication stochastic kriging |
title | hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R |
title_full | hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R |
title_fullStr | hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R |
title_full_unstemmed | hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R |
title_short | hetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R |
title_sort | hetgp heteroskedastic gaussian process modeling and sequential design in r |
topic | input-dependent noise level-set estimation optimization replication stochastic kriging |
url | https://www.jstatsoft.org/index.php/jss/article/view/3655 |
work_keys_str_mv | AT mickaelbinois hetgpheteroskedasticgaussianprocessmodelingandsequentialdesigninr AT robertbgramacy hetgpheteroskedasticgaussianprocessmodelingandsequentialdesigninr |