VMPF: Variance Minimization of the Predicted Front
Description
Executes nsteps
iterations of the VMPF algorithm to an object of class
mkm
. At each step, a multiobjective kriging model is reestimated
(including covariance parameters reestimation).
Usage
1 2 
Arguments
model 
An object of class 
fun 
The multiobjective and constraint cost function to be optimized.
This function must return a vector with the size of 
nsteps 
An integer representing the desired number of iterations, 
lower 
Vector of lower bounds for the variables to be optimized over
(default: 0 with length 
upper 
Vector of upper bounds for the variables to be optimized over
(default: 1 with length 
quiet 
Logical indicating the verbosity of the routine, 
control 
An optional list of control parameters that controlls the optimization algorithm. One can control:

modelcontrol 
An optional list of control parameters to the

Details
The infill point is sampled from the most uncertain design of a predicted Pareto set. This set is predicted using nsga2 algorithm and the mean value of the mkm predictor.
Value
an updated object of class mkm
.
Examples
1 2 3 4 5 6 7 8 9 10 11 12 13  # 
# The Nowacki Beam
# 
n < 20
d < 2
nsteps < 2 # value has been set to 2 to save compliation time, change this value to 40.
fun < nowacki_beam
doe < replicate(d,sample(0:n,n))/n
res < t(apply(doe, 1, fun))
model < mkm(doe, res, modelcontrol = list(objective = 1:2, lower = rep(0.1,d)))
model < VMPF(model, fun, nsteps, quiet = FALSE)
plot(nowacki_beam_tps$set)
points(ps(model@response[which(model@feasible),model@objective])$set, col = 'green', pch = 19)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker. Vote for new features on Trello.