Surrogate Model Optimization Toolbox

Surrogate model optimization algorithm for computationally expensive global optimization problems

Description: Surrogate model toolbox for
- unconstrained continuous
- constrained integer
- constrained mixed-integer
global optimization problems that are computationally expensive.
The user can choose beween different options for
- the surrogate model
- the sampling strategy
- the initial experimental design
The user can determine the maximum number of allowed function evaluations, the number of points in the initial starting design, and one or more points that are added to the starting design.
The algorithm is useful when a single function evaluation is very time-consuming and therefore good approximations of the global optimum must be found within a very restricted number of evaluations.