Choosing the Sample Size of a Computer Experiment: A Practical Guide
By Jason L. Loeppky, J. Sacks, William J. Welch. 2009
684 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computational Theory and Mathematics, Management Science and Operations Research
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computational Theory and Mathematics
- Yew-Soon Ong
- Jianfei Cai
- Haitao Liu
- Victor Picheny
- David Ginsbourger
- Emmanuel Vazquez
- Julien Bect
- Wei Gong
In Management Science and Operations Research
- Victor Picheny
- David Ginsbourger
- Yew-Soon Ong
- Jianfei Cai
- Haitao Liu
- Emmanuel Vazquez
- Bela Gipp
- Julien Bect
Citations in the corpus
- A comprehensive evaluation of various sensitivity analysis methods: A case study with a hydrological model (2014, background): “, 1989) and how many sample points are sufficient for the experiment (Loeppky et al., 2009).”
- Remarks on multi-output Gaussian process regression (2018, background): “[91] has provided reasons and evidence supporting that this initial training size is able to build an effective initial GP model.”
- Arbitrary polynomial chaos expansion method for uncertainty quantification and global sensitivity analysis in structural dynamics (2020, methodology): “For computer experiment, the initial 390 number of training data points is commonly set to 10d (d=5 is the number of model inputs) 391 based on the well-known “n=10d” rule of thumb [59, 60].”
- Research-paper recommender systems: a literature survey (2015, background): “being displayed, and the variations in the results [271,272].”
- Sequential design of computer experiments for the estimation of a probability of failure (2010, methodology): “Several strategies of this kind have been proposed recently in the literature by Ranjan et al. (2008), Bichon et al. (2008), Picheny et al.”
- A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design (2017, methodology): “It is worth noting that when using a GP model for approximation, Loeppky et al. (2009) showed that the appropriate initial sample size should be ten times the dimensionality of the target function.”