2021
DOI: 10.48550/arxiv.2105.04599
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Budget-limited distribution learning in multifidelity problems

Abstract: Multifidelity methods are widely used for statistical estimation of quantities of interest (QoIs) in uncertainty quantification using simulation codes of differing costs and accuracies. Many methods approximate numerical-valued statistics that represent only limited information of the QoIs. In this paper, we introduce a semi-parametric approach that aims to effectively describe the distribution of a scalar-valued QoI in the multifidelity setup. Under a linear model hypothesis, we propose an exploration-exploit… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2022
2022
2022
2022

Publication Types

Select...
1

Relationship

0
1

Authors

Journals

citations
Cited by 1 publication
(1 citation statement)
references
References 31 publications
0
1
0
Order By: Relevance
“…Fletcher, McNally, and Virgin (2021) and have shown that multifidelity training is feasible and can lead to improvements in accuracy for CAM4 and CAM5 examples. More recent approaches by Liu, Pareschi, and Zhu (2021) and Xu and Narayan (2021) seek to find efficient and mathematically rigorous ways of learning a surrogate model with bi-or multifidelity training sources, but these approaches have not yet been tested on global climate models.…”
Section: Multifidelity / Multiresolution Modelingmentioning
confidence: 99%
“…Fletcher, McNally, and Virgin (2021) and have shown that multifidelity training is feasible and can lead to improvements in accuracy for CAM4 and CAM5 examples. More recent approaches by Liu, Pareschi, and Zhu (2021) and Xu and Narayan (2021) seek to find efficient and mathematically rigorous ways of learning a surrogate model with bi-or multifidelity training sources, but these approaches have not yet been tested on global climate models.…”
Section: Multifidelity / Multiresolution Modelingmentioning
confidence: 99%