Gaussian process regression models, a.k.a. Kriging models, are applied to global multi-objective optimization of black-box functions. Multi-objective Expected Improvement and Step-wise Uncertainty Reduction sequential infill criteria are available. A quantification of uncertainty on Pareto fronts is provided using conditional simulations.
| Version: |
1.1.0 |
| Depends: |
DiceKriging (≥ 1.5.3), emoa, methods |
| Imports: |
Rcpp (≥ 0.11.1), rgenoud, pbivnorm, pso, randtoolbox, KrigInv, MASS, DiceDesign (≥ 1.4), ks |
| LinkingTo: |
Rcpp |
| Suggests: |
knitr |
| Published: |
2017-06-29 |
| Author: |
Mickael Binois, Victor Picheny |
| Maintainer: |
Mickael Binois <mickael.binois at chicagobooth.edu> |
| License: |
GPL-3 |
| NeedsCompilation: |
yes |
| Materials: |
NEWS |
| CRAN checks: |
GPareto results |