The bayesopt function performs all Bayesian optimization. The bayesopt algorithm is described here. When finding the best parameters of an SVM using fitcsvm, then according to the documentation of the OptimizeHyperparameters option:
"The optimization attempts to minimize the cross-validation loss (error) for fitcsvm by varying the parameters. For information about cross-validation loss see Classification Loss." In other words, the objective function that bayesopt tries to minimize is the cross-validation loss.
I hope that is enough detail.
Alan Weiss
MATLAB mathematical toolbox documentation
