Gaussian Processes for Pareto Front Estimation and Optimization
Package GPareto
Sequential multi-objective Expected Improvement maximization and model...
Function to build integration points (for the SUR criterion)
Non-dominated points with respect to a reference
Estimation of Pareto set density
Plot uncertainty
Plot multi-objective optimization results and post-processing
Pareto front visualization
Visualisation of Pareto front and set
Display the Symmetric Deviation Function
Symmetrical difference of RNP sets
Predict function for list of km models.
Prevention of numerical instability for a new observation
Conditional Pareto Front simulations
Expected Hypervolume Improvement with m objectives
Expected Maximin Improvement with m objectives
Maximization of multiobjective infill criterion
Batch Expected Hypervolume Improvement with m objectives
Analytical expression of the SMS-EGO criterion with m>1 objectives
Analytical expression of the SUR criterion for two or three objectives...
EGO algorithm for multiobjective optimization
Class for fast to compute objective.
Fast-to-evaluate function wrapper
Get design corresponding to an objective target
Test functions of x
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.