View in EDS

Nested polynomial trends for the improvement of Gaussian process-based predictors.

Saved in:
Bibliographic Details
Title: Nested polynomial trends for the improvement of Gaussian process-based predictors.
Authors: Perrin, G.1 guillaume.perrin2@cea.fr, Soize, C.2, Marque-Pucheu, S.1,3, Garnier, J.3
Source: Journal of Computational Physics. Oct2017, Vol. 346, p389-402. 14p.
Subjects: Polynomials, Gaussian processes, Sensitivity analysis, Mathematical mappings, Parameterization
Abstract: The role of simulation keeps increasing for the sensitivity analysis and the uncertainty quantification of complex systems. Such numerical procedures are generally based on the processing of a huge amount of code evaluations. When the computational cost associated with one particular evaluation of the code is high, such direct approaches based on the computer code only, are not affordable. Surrogate models have therefore to be introduced to interpolate the information given by a fixed set of code evaluations to the whole input space. When confronted to deterministic mappings, the Gaussian process regression (GPR), or kriging, presents a good compromise between complexity, efficiency and error control. Such a method considers the quantity of interest of the system as a particular realization of a Gaussian stochastic process, whose mean and covariance functions have to be identified from the available code evaluations. In this context, this work proposes an innovative parametrization of this mean function, which is based on the composition of two polynomials. This approach is particularly relevant for the approximation of strongly non linear quantities of interest from very little information. After presenting the theoretical basis of this method, this work compares its efficiency to alternative approaches on a series of examples. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Computational Physics is the property of Academic Press Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 124113884
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Nested polynomial trends for the improvement of Gaussian process-based predictors.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Perrin%2C+G%2E%22">Perrin, G.</searchLink><relatesTo>1</relatesTo><i> guillaume.perrin2@cea.fr</i><br /><searchLink fieldCode="AR" term="%22Soize%2C+C%2E%22">Soize, C.</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Marque-Pucheu%2C+S%2E%22">Marque-Pucheu, S.</searchLink><relatesTo>1,3</relatesTo><br /><searchLink fieldCode="AR" term="%22Garnier%2C+J%2E%22">Garnier, J.</searchLink><relatesTo>3</relatesTo>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Journal+of+Computational+Physics%22">Journal of Computational Physics</searchLink>. Oct2017, Vol. 346, p389-402. 14p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Polynomials%22">Polynomials</searchLink><br /><searchLink fieldCode="DE" term="%22Gaussian+processes%22">Gaussian processes</searchLink><br /><searchLink fieldCode="DE" term="%22Sensitivity+analysis%22">Sensitivity analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Mathematical+mappings%22">Mathematical mappings</searchLink><br /><searchLink fieldCode="DE" term="%22Parameterization%22">Parameterization</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The role of simulation keeps increasing for the sensitivity analysis and the uncertainty quantification of complex systems. Such numerical procedures are generally based on the processing of a huge amount of code evaluations. When the computational cost associated with one particular evaluation of the code is high, such direct approaches based on the computer code only, are not affordable. Surrogate models have therefore to be introduced to interpolate the information given by a fixed set of code evaluations to the whole input space. When confronted to deterministic mappings, the Gaussian process regression (GPR), or kriging, presents a good compromise between complexity, efficiency and error control. Such a method considers the quantity of interest of the system as a particular realization of a Gaussian stochastic process, whose mean and covariance functions have to be identified from the available code evaluations. In this context, this work proposes an innovative parametrization of this mean function, which is based on the composition of two polynomials. This approach is particularly relevant for the approximation of strongly non linear quantities of interest from very little information. After presenting the theoretical basis of this method, this work compares its efficiency to alternative approaches on a series of examples. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Journal of Computational Physics is the property of Academic Press Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=124113884
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.jcp.2017.05.051
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 389
    Subjects:
      – SubjectFull: Polynomials
        Type: general
      – SubjectFull: Gaussian processes
        Type: general
      – SubjectFull: Sensitivity analysis
        Type: general
      – SubjectFull: Mathematical mappings
        Type: general
      – SubjectFull: Parameterization
        Type: general
    Titles:
      – TitleFull: Nested polynomial trends for the improvement of Gaussian process-based predictors.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Perrin, G.
      – PersonEntity:
          Name:
            NameFull: Soize, C.
      – PersonEntity:
          Name:
            NameFull: Marque-Pucheu, S.
      – PersonEntity:
          Name:
            NameFull: Garnier, J.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 10
              Text: Oct2017
              Type: published
              Y: 2017
          Identifiers:
            – Type: issn-print
              Value: 00219991
          Numbering:
            – Type: volume
              Value: 346
          Titles:
            – TitleFull: Journal of Computational Physics
              Type: main
ResultId 1