Kriging-based optimization for computer experiments

[Up] [Top]

Documentation for package ‘DiceOptim’ version 1.3.3

Help Pages

AEI Augmented Expected Improvement
AEI.grad AEI's Gradient
AEI.grad_optim AEI's Gradient
AKG Approximate Knowledge Gradient (AKG)
AKG.grad AKG's Gradient
AKG.grad_optim AKG's Gradient, for usage in max_AKG
CL.nsteps Parallelized version of EGO.nsteps, based on the CL strategy
DiceOptim Kriging-based optimization methods for computer experiments
EGO.nsteps Sequential EI maximization and model re-estimation, with a number of iterations fixed in advance by the user
EI Analytical expression of the Expected Improvement criterion (noise-free version)
EI.grad Analytical gradient of the Expected Improvement criterion (noise-free version)
EI.plugin Expected Improvement with plugin
EI.plugin.grad Analytical gradient of the Expected Improvement criterion with plug-in
EQI Expected Quantile Improvement
EQI.grad EQI's Gradient
EQI.grad_optim EQI's Gradient, for usage in max_EQI
kriging.quantile Kriging quantile
kriging.quantile.grad Analytical gradient of the Kriging quantile of level alpha
kriging.quantile.grad_optim Analytical gradient of the Kriging quantile of level alpha
max_AEI Maximizer of the Augmented Expected Improvement criterion function
max_AKG Maximizer of the Expected Quantile Improvement criterion function
max_EI One-shot maximization of the Expected Improvement criterion (noise-free version)
max_EI.plugin Maximizer of the Expected Improvement criterion function with plugin of the minimum.
max_EQI Maximizer of the Expected Quantile Improvement criterion function
max_qEI.CL One-shot pseudo-maximization of qEI using the Constant Liar strategy
min_quantile Minimizer of the Kriging quantile.
noisy.optimizer Optimization of homogenously noisy functions based on Kriging
qEI Monte-Carlo estimation of the multipoints Expected Improvement criterion (noise-free version)
update_km Update of a Kriging model when adding new observation