vorob_optim_parallel2: Parallel Vorob'ev criterion

In KrigInv: Kriging-Based Inversion for Deterministic and Noisy Computer Experiments

Parallel Vorob'ev criterion

Description

Evaluation of the Vorob'ev criterion for some candidate points, assuming that some other points are also going to be evaluated. To be used in optimization routines, like in max_vorob_parallel. To avoid numerical instabilities, the new points are evaluated only if they are not too close to an existing observation, or if there is some observation noise. The criterion is the integral of the posterior Vorob'ev uncertainty.

Usage

vorob_optim_parallel2(x, other.points,
integration.points, integration.weights = NULL,
intpoints.oldmean, intpoints.oldsd, precalc.data,
model, T, new.noise.var = NULL,
batchsize, alpha, current.vorob,
penalisation = NULL, typeEx = ">")

Arguments

x	Input vector of size d at which one wants to evaluate the criterion. This argument corresponds to only ONE point.
other.points	Vector giving the other batchsize-1 points at which one wants to evaluate the criterion
integration.points	p*d matrix of points for numerical integration in the X space.
integration.weights	Vector of size p corresponding to the weights of these integration points.
intpoints.oldmean	Vector of size p corresponding to the kriging mean at the integration points before adding x to the design of experiments.
intpoints.oldsd	Vector of size p corresponding to the kriging standard deviation at the integration points before adding x to the design of experiments.
precalc.data	List containing useful data to compute quickly the updated kriging variance. This list can be generated using the precomputeUpdateData function.
model	Object of class km (Kriging model).
T	Target value (scalar). The criterion CANNOT be used with multiple thresholds.
new.noise.var	Optional scalar value of the noise variance of the new observations.
batchsize	Number of points to sample simultaneously. The sampling criterion will return batchsize points at a time for sampling.
alpha	The Vorob'ev threshold.
current.vorob	Current value of the vorob criterion (before adding new observations)
penalisation	Optional penalization constant for type I errors. If equal to zero, computes the Type II criterion.
typeEx	A character (">" or "<") identifying the type of excursion

Details

The first argument x has been chosen to be a vector of size d so that an optimizer like genoud can optimize it easily. The second argument other.points is a vector of size (batchsize-1)*d corresponding to the batchsize-1 other points.

Value

Parallel Vorob'ev value

Author(s)

Clement Chevalier (University of Neuchatel, Switzerland)

Dario Azzimonti (IDSIA, Switzerland)

References

Chevalier C., Ginsbouger D., Bect J., Molchanov I. (2013) Estimating and quantifying uncertainties on level sets using the Vorob'ev expectation and deviation with gaussian process models mODa 10, Advances in Model-Oriented Design and Analysis, Contributions to Statistics, pp 35-43

Chevalier C. (2013) Fast uncertainty reduction strategies relying on Gaussian process models Ph.D Thesis, University of Bern

See Also

EGIparallel, max_vorob_parallel

Examples

#vorob_optim_parallel2

set.seed(9)
N <- 20 #number of observations
T <- 80 #threshold
testfun <- branin

#a 20 points initial design
design <- data.frame( matrix(runif(2*N),ncol=2) )
response <- testfun(design)

#km object with matern3_2 covariance
#params estimated by ML from the observations
model <- km(formula=~., design = design,
	response = response,covtype="matern3_2")

###we need to compute some additional arguments:
#integration points, and current kriging means and variances at these points
integcontrol <- list(n.points=50,distrib="vorob",init.distrib="MC")
obj <- integration_design(integcontrol=integcontrol,
lower=c(0,0),upper=c(1,1),model=model,T=T)

integration.points <- obj$integration.points
integration.weights <- obj$integration.weights
alpha <- obj$alpha
pred <- predict_nobias_km(object=model,newdata=integration.points,
                          type="UK",se.compute=TRUE)
intpoints.oldmean <- pred$mean ; intpoints.oldsd<-pred$sd

#another precomputation
precalc.data <- precomputeUpdateData(model,integration.points)

batchsize <- 4
other.points <- c(0.7,0.5,0.5,0.9,0.9,0.8)
x <- c(0.1,0.2)
#one evaluation of the vorob_optim_parallel criterion2
#we calculate the expectation of the future "vorob" uncertainty when
#1+3 points are added to the doe
#the 1+3 points are (0.1,0.2) and (0.7,0.5), (0.5,0.9), (0.9,0.8)
vorob_optim_parallel2(x=x,other.points,integration.points=integration.points,
          integration.weights=integration.weights,
          intpoints.oldmean=intpoints.oldmean,intpoints.oldsd=intpoints.oldsd,
          precalc.data=precalc.data,T=T,model=model,
          batchsize=batchsize,alpha=alpha,current.vorob=Inf)


n.grid <- 20 #you can run it with 100
x.grid <- y.grid <- seq(0,1,length=n.grid)
x <- expand.grid(x.grid, y.grid)
vorob_parallel.grid <- apply(X=x,FUN=vorob_optim_parallel2,MARGIN=1,other.points,
          integration.points=integration.points,
          integration.weights=integration.weights,
          intpoints.oldmean=intpoints.oldmean,intpoints.oldsd=intpoints.oldsd,
          precalc.data=precalc.data,T=T,model=model,
          batchsize=batchsize,alpha=alpha,current.vorob=Inf)
z.grid <- matrix(vorob_parallel.grid, n.grid, n.grid)

#plots: contour of the criterion, doe points and new point
image(x=x.grid,y=y.grid,z=z.grid,col=grey.colors(10))
contour(x=x.grid,y=y.grid,z=z.grid,15,add=TRUE)
points(design, col="black", pch=17, lwd=4,cex=2)
points(matrix(other.points,ncol=2,byrow=TRUE), col="red", pch=17, lwd=4,cex=2)

i.best <- which.min(vorob_parallel.grid)
points(x[i.best,], col="blue", pch=17, lwd=4,cex=3)

#plots the real (unknown in practice) curve f(x)=T
testfun.grid <- apply(x,1,testfun)
z.grid.2 <- matrix(testfun.grid, n.grid, n.grid)
contour(x.grid,y.grid,z.grid.2,levels=T,col="blue",add=TRUE,lwd=5)
title("Contour lines of vorob_parallel criterion (black) and of f(x)=T (blue)")

KrigInv documentation built on Sept. 9, 2022, 5:08 p.m.