|
Abstract:
|
Computer experiments often require dense sweeps over input parameters to obtain a qualitative understanding of their response. However, such sweeps are unnecessary in regions where the response is easily predicted; well-chosen designs could allow a mapping of the response with far fewer simulation runs. Thus, there is a need for computationally inexpensive surrogate models and an accompanying method for selecting small designs. I explore a semiparametric nonstationary modeling methodology for addressing this need that couples stationary Gaussian processes with treed partitioning. A Bayesian perspective yields an explicit measure of (nonstationary) predictive uncertainty that can be used to guide sampling. The benefit of adaptive sampling is illustrated through several examples, including a motivating example which involves the computational fluid dynamics of a NASA re-entry vehicle.
|
