GPareto: Gaussian Processes for Pareto Front Estimation and Optimization
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: |
1.1.1 |
| Depends: |
DiceKriging, emoa |
| Imports: |
Rcpp (≥ 0.12.15), methods, rgenoud, pbivnorm, pso, randtoolbox, KrigInv, MASS, DiceDesign, ks, rgl |
| LinkingTo: |
Rcpp |
| Suggests: |
knitr |
| Published: |
2018-01-31 |
| Author: |
Mickael Binois, Victor Picheny |
| Maintainer: |
Mickael Binois <mbinois at mcs.anl.gov> |
| License: |
GPL-3 |
| NeedsCompilation: |
yes |
| Materials: |
NEWS |
| CRAN checks: |
GPareto results |
Downloads:
Reverse dependencies:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=GPareto
to link to this page.