2015
DOI: 10.1016/j.ifacol.2015.12.183
|View full text |Cite
|
Sign up to set email alerts
|

A novel sequential exploration-exploitation sampling strategy for global metamodeling

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
17
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 22 publications
(17 citation statements)
references
References 11 publications
0
17
0
Order By: Relevance
“…7b 2 . For instance, an error at any point x can be approximated as the superposition of the relative errors between the current metamodel and the leave-one-out metamodels considering successively the lack of each sample [45], as follows The method has also been adopted by Kim et al [50] and extended as a weighted version in Jiang et al [42].…”
Section: Continuousmentioning
confidence: 99%
See 1 more Smart Citation
“…7b 2 . For instance, an error at any point x can be approximated as the superposition of the relative errors between the current metamodel and the leave-one-out metamodels considering successively the lack of each sample [45], as follows The method has also been adopted by Kim et al [50] and extended as a weighted version in Jiang et al [42].…”
Section: Continuousmentioning
confidence: 99%
“…proportional to the average minimum distance of all sample points. A similar approach designed by Jiang et al [42] defines the cluster threshold S Jiang as detailed in Box 1.…”
Section: Continuous Distance Criterionmentioning
confidence: 99%
“…The query-by-committee (QBC) strategy [ 31 , 35 , 36 ] uses the predictions of multiple competing surrogate models as a committee to predict the response at a candidate point. The cross validation (CV) based adaptive sampling [ 37 , 38 , 39 ] estimates the prediction error at a candidate point using a cross-validation process. Finally, The gradient-based adaptive sampling [ 40 , 41 , 42 ] uses the local gradient information of the model to represent the prediction errors.…”
Section: Related Workmentioning
confidence: 99%
“…The main limitation of the current adaptive sampling methods is that they seek to identify the points that most favor a predetermined meta-model based on its performance [ 37 , 38 , 39 , 40 , 41 , 42 ]. However, it is not easy to determine if the points were selected due to the target system behavior or to an intrinsic limitation of the selected meta-model.…”
Section: Related Workmentioning
confidence: 99%
“…Another sampling challenge touched on above is efficient sampling for updating surrogate models, balancing exploration (global search) and exploitation (precise solution) of the design space [33,34,35]. Previous studies aimed to achieve the following goals simultaneously: 1) finding a global solution, 2) finding an accurate local solution, and 3) limiting the number of high-fidelity simulations.…”
Section: Introductionmentioning
confidence: 99%