1Blurb::
2Use the surrogate based optimization method
3Description::
4A surrogate-based optimization method will be used.  The surrogate
5employed in \c sbo is a Gaussian process surrogate.
6
7The main difference between \c ego and the
8\c sbo approach is the objective function being optimized.
9\c ego relies on an expected improvement function, while in
10\c sbo, the optimization proceeds using an evolutionary
11algorithm (\ref method-coliny_ea) on the Gaussian process surrogate:
12it is a standard surrogate-based optimization. Also note that the \c
13sbo option can support optimization over discrete variables (the
14discrete variables are relaxed) while \c ego cannot.
15
16This is not the same as \ref method-surrogate_based_global.
17
18Topics::
19Examples::
20Theory::
21Faq::
22See_Also::
23