- "Classical" RSMs, but only as a jumping-off point.
The interplay between mathematical models, numerical approximation, simulation, computer experiments, and (field) data.
- Gaussian process (GP) spatial models, emphasizing
- surrogate computer modeling,
- sequential design, Bayesian optimization,
- calibration,
- variable selection and sensitivity analysis, and more.
- Uncertainty quantification, where statistics ought to monopolize but sometimes doesn't.
- Machine learning methods:
- "big-\(n\)" GP solutions (sparsity), non-stationary GP modeling, the frontier …