Learn R Programming

DiceOptim (version 2.1.1)

Kriging-Based Optimization for Computer Experiments

Description

Efficient Global Optimization (EGO) algorithm as described in "Roustant et al. (2012)" and adaptations for problems with noise ("Picheny and Ginsbourger, 2012") , parallel infill, and problems with constraints.

Copy Link

Version

Install

install.packages('DiceOptim')

Monthly Downloads

77

Version

2.1.1

License

GPL-2 | GPL-3

Maintainer

V. Picheny

Last Published

February 2nd, 2021

Functions in DiceOptim (2.1.1)

EQI

Expected Quantile Improvement
AEI.grad

AEI's Gradient
AKG

Approximate Knowledge Gradient (AKG)
AEI

Augmented Expected Improvement
AKG.grad

AKG's Gradient
DiceOptim-package

Kriging-based optimization methods for computer experiments
EGO.cst

Sequential constrained Expected Improvement maximization and model re-estimation, with a number of iterations fixed in advance by the user
EI

Analytical expression of the Expected Improvement criterion
EGO.nsteps

Sequential EI maximization and model re-estimation, with a number of iterations fixed in advance by the user
EI.grad

Analytical gradient of the Expected Improvement criterion
EQI.grad

EQI's Gradient
checkPredict

Prevention of numerical instability for a new observation
fastfun-class

Class for fast to compute objective.
ParrConstraint

2D constraint function
fastfun

Fastfun function
kriging.quantile.grad

Analytical gradient of the Kriging quantile of level beta
max_AEI

Maximizer of the Augmented Expected Improvement criterion function
critcst_optimizer

Maximization of constrained Expected Improvement criteria
easyEGO

User-friendly wrapper of the functions fastEGO.nsteps and TREGO.nsteps. Generates initial DOEs and kriging models (objects of class km), and executes nsteps iterations of either EGO or TREGO.
integration_design_cst

Generic function to build integration points (for the SUR criterion)
kriging.quantile

Kriging quantile
crit_AL

Expected Augmented Lagrangian Improvement
goldsteinprice

2D test function
update_km_noisyEGO

Update of one or two Kriging models when adding new observation
max_AKG

Maximizer of the Expected Quantile Improvement criterion function
TREGO.nsteps

Trust-region based EGO algorithm.
hartman4

4D test function
easyEGO.cst

EGO algorithm with constraints
max_EI

Maximization of the Expected Improvement criterion
branin2

2D test function
qEI

Analytical expression of the multipoint expected improvement (qEI) criterion
test_feas_vec

Test constraints violation (vectorized)
max_EQI

Maximizer of the Expected Quantile Improvement criterion function
max_crit

Maximization of the Expected Improvement criterion
qEI.grad

Gradient of the multipoint expected improvement (qEI) criterion
crit_EFI

Expected Feasible Improvement
crit_SUR_cst

Stepwise Uncertainty Reduction criterion
noisy.optimizer

Optimization of homogenously noisy functions based on Kriging
sphere6

6D sphere function
qEGO.nsteps

Sequential multipoint Expected improvement (qEI) maximizations and model re-estimation
fastEGO.nsteps

Sequential EI maximization and model re-estimation, with a number of iterations fixed in advance by the user
max_qEI

Maximization of multipoint expected improvement criterion (qEI)
min_quantile

Minimization of the Kriging quantile.
rosenbrock4

4D test function
sampleFromEI

Sampling points according to the expected improvement criterion