Parameters of the GP surrogate models. The maximum distance $rmaxA$ is derived using the Mahalanobis distance given by the covariance matrix $σ2C$. $Qχ2(0.99,D)$ is the 0.99-quantile of the $χD2$ distribution, and therefore $Qχ2(0.99,D)$ is the 0.99-quantile of the norm of a $D$-dimensional normal distributed random vector.
training set selection method $TSS$ (TSS1), (TSS2), (TSS3), (TSS4)
maximum distance $rmaxA$ $2Qχ2(0.99,D)$, $4Qχ2(0.99,D)$
$Nmax$ $10·D$, $15·D$, $20·D$
covariance function $K$ $KSE$, $KMate´rnν=3/2$, $KMate´rnν=5/2$
training set selection method $TSS$ (TSS1), (TSS2), (TSS3), (TSS4)
maximum distance $rmaxA$ $2Qχ2(0.99,D)$, $4Qχ2(0.99,D)$
$Nmax$ $10·D$, $15·D$, $20·D$
covariance function $K$ $KSE$, $KMate´rnν=3/2$, $KMate´rnν=5/2$