<b>hetGP</b>: Heteroskedastic Gaussian Process Modeling and Sequential Design in <i>R</i>

An increasing number of time-consuming simulators exhibit a complex noise structure that depends on the inputs. For conducting studies with limited budgets evaluations, new surrogate methods are required in order to simultaneously model mean and variance fields. To this end, we present hetGP package, implementing many recent advances Gaussian process modeling input-dependent noise. First, describe simple, yet efficient, joint framework relies replication for both speed accuracy. Then tackle issue data acquisition leveraging exploration sequential manner various goals, such as obtaining globally accurate model, optimization, or contour finding. Reproducible illustrations provided throughout.