Improving kriging surrogates of high-dimensional design models by Partial Least Squares dimension reduction - Archive ouverte HAL Access content directly
Journal Articles Structural and Multidisciplinary Optimization Year : 2016

Improving kriging surrogates of high-dimensional design models by Partial Least Squares dimension reduction

Abstract

Engineering computer codes are often compu- tationally expensive. To lighten this load, we exploit new covariance kernels to replace computationally expensive codes with surrogate models. For input spaces with large dimensions, using the kriging model in the standard way is computationally expensive because a large covariance matrix must be inverted several times to estimate the param- eters of the model. We address this issue herein by con- structing a covariance kernel that depends on only a few parameters. The new kernel is constructed based on infor- mation obtained from the Partial Least Squares method. Promising results are obtained for numerical examples with up to 100 dimensions, and significant computational gain is obtained while maintaining sufficient accuracy.

Domains

Other
Fichier principal
Vignette du fichier
Bouhlel_18187..pdf (2.16 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01598259 , version 1 (04-10-2017)
hal-01598259 , version 2 (19-04-2021)

Identifiers

Cite

Mohamed Amine Bouhlel, Nathalie Bartoli, Abdelkader Otsmane, Joseph Morlier. Improving kriging surrogates of high-dimensional design models by Partial Least Squares dimension reduction. Structural and Multidisciplinary Optimization, 2016, vol. 53 (n° 5), pp. 935-952. ⟨10.1007/s00158-015-1395-9⟩. ⟨hal-01598259v2⟩
283 View
1234 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More