Local Optimization of Black-Box Function with High or Infinite-Dimensional Inputs: Application to Nuclear Safety

Abstract : An adaptation of Response Surface Methodology (RSM) when the covariate is of high or infinite dimensional is proposed, providing a tool for black-box optimization in this context. We combine dimension reduction techniques with classical multivariate Design of Experiments (DoE). We propose a method to generate experimental designs and extend usual properties (orthogonality, rotatability,...) of multivariate designs to general high or infinite dimensional contexts. Different dimension reduction basis are considered (including data-driven basis). The methodology is illustrated on simulated functional data and we discuss the choice of the different parameters, in particular the dimension of the approximation space. The method is finally applied to a problem of nuclear safety.
Type de document :
Pré-publication, Document de travail
MAP5 2015-11. 2015
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01144628
Contributeur : Angelina Roche <>
Soumis le : mardi 17 novembre 2015 - 16:54:29
Dernière modification le : mardi 11 octobre 2016 - 15:17:07

Fichiers

FuncSurResp_HaL.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01144628, version 3
  • ARXIV : 1506.02886

Collections

Citation

Angelina Roche. Local Optimization of Black-Box Function with High or Infinite-Dimensional Inputs: Application to Nuclear Safety. MAP5 2015-11. 2015. <hal-01144628v3>

Partager

Métriques

Consultations de
la notice

120

Téléchargements du document

76