Noisy kriging-based optimization methods: A unified implementation within the DiceOptim package

Picheny, Victor; Ginsbourger, David

doi:10.1016/j.csda.2013.03.018

Cited by 40 publications

(35 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The maximum and minimum values of τ ( x ) are linked to R f , i.e., the range of the objective value in the region of interest (as in Huang et al, 2006 andGinsbourger, 2014 ). With light noise, τ ( x ) varies between 15% and 60% of R f .…”

Section: Algorithmic Factorsmentioning

confidence: 99%

“…The idea behind the correlated knowledge-gradient (CKG) infill criterion is that in noisy environments, the Kriging prediction ˆ f (x ) may be closer to f ( x ) than the sample mean f (x ) ; therefore, points are selected based on their effect on the Kriging prediction ( Frazier et al, 2009;Picheny & Ginsbourger, 2014 ). More specifically, at iteration k + 1 , the improvement that would result from sampling alternative x k +1 is defined as:…”

Section: Correlated Knowledge-gradient (Ckg)mentioning

confidence: 99%

“…As most of the algorithms allow for revisits, the covariance matrix of the metamodel may become ill-conditioned; we follow the guidelines in Section 4.3 of Picheny and Ginsbourger (2014) for handling this issue. Table 7 shows the average running time of different algorithms for one macroreplication in all the test problems (on a Dell desktop with 3.4 gigahertz Core i7-2600 processor and 8 gigabytes of RAM).…”

Section: Implementation Detailsmentioning

confidence: 99%

“…is the inverse CDF of standard normal distribution and β ∈ (0, 0.5] in MQ and β ∈ [0.5, 1] in SKO and EQI ( Picheny & Ginsbourger, 2014 ).…”

Section: Mq Algorithmmentioning

confidence: 99%

“…To fit the stochastic Kriging models, we used the code provided on http://stochastickriging.net/ , for CKG we used the code on http://people.orie.cornell.edu/pfrazier/src.html , the TSSO code was based on http://www.ise.nus.edu.sg/staff/ngsh/download/ matlabdocs/SOK/ and the eTSSO code was obtained from the authors of Liu et al (2014) . The remaining algorithms were coded by the authors themselves and compared with the DiceOptim package in R ( Picheny & Ginsbourger, 2014 ) to ensure correctness.…”

Section: Implementation Detailsmentioning

confidence: 99%

See 4 more Smart Citations

Comparison of Kriging-Based Methods for Simulation Optimization with Heterogeneous Noise

2015

Self Cite

View full text Add to dashboard Cite

a b s t r a c tIn this article we investigate the unconstrained optimization (minimization) of the performance of a system that is modeled through a discrete-event simulation. In recent years, several algorithms have been proposed which extend the traditional Kriging-based simulation optimization algorithms (assuming deterministic outputs) to problems with noise. Our objective in this paper is to compare the relative performance of a number of these algorithms on a set of well-known analytical test functions, assuming different patterns of heterogeneous noise. We also apply the algorithms to a popular inventory test problem. The conclusions and insights obtained may serve as a useful guideline for researchers aiming to apply Kriging-based algorithms to solve engineering and/or business problems, and may be useful in the development of future algorithms.

show abstract

Section: Algorithmic Factorsmentioning

confidence: 99%

Section: Correlated Knowledge-gradient (Ckg)mentioning

confidence: 99%

Section: Implementation Detailsmentioning

confidence: 99%

“…is the inverse CDF of standard normal distribution and β ∈ (0, 0.5] in MQ and β ∈ [0.5, 1] in SKO and EQI ( Picheny & Ginsbourger, 2014 ).…”

Section: Mq Algorithmmentioning

confidence: 99%

Section: Implementation Detailsmentioning

confidence: 99%

See 3 more Smart Citations

Comparison of Kriging-Based Methods for Simulation Optimization with Heterogeneous Noise

2015

Self Cite

View full text Add to dashboard Cite

show abstract

Optimal robust and tolerance design for computer experiments with mixture proportion inputs

Han

Tan

2017

Quality & Reliability Eng

View full text Add to dashboard Cite

Computer experiments often have inputs that are proportions/fractions of components in a mixture. In these mixture computer experiments, it can be of interest to perform robust and tolerance design on the mixture proportions since the proportions are subjected to noise variations. Traditionally, manufacturing of mixture products is controlled via interval tolerances for mixture amounts. In this paper, an optimal tolerance region for proportions, which gives optimal quality cost among all possible tolerance regions for mixture proportions with the same acceptance probability, is proposed for integrated parameter and tolerance design in mixture computer experiments. Real examples are given to demonstrate the improvements that can be achieved with the optimal tolerance region.

show abstract

Overview of Problem Formulations and Optimization Algorithms in the Presence of Uncertainty

Balesdent

Brevault

Mitard

et al. 2020

Springer Optimization and Its Applications

View full text Add to dashboard Cite

Noisy kriging-based optimization methods: A unified implementation within the DiceOptim package

Cited by 40 publications

References 27 publications

Comparison of Kriging-Based Methods for Simulation Optimization with Heterogeneous Noise

Comparison of Kriging-Based Methods for Simulation Optimization with Heterogeneous Noise

Optimal robust and tolerance design for computer experiments with mixture proportion inputs

Overview of Problem Formulations and Optimization Algorithms in the Presence of Uncertainty

Contact Info

Product

Resources

About