Option to set Random forest as Surrogate benchmark model

yanyachen / rBayesianOptimization

Bayesian Optimization of Hyperparameters

84 stars 21 forks source link

Option to set Random forest as Surrogate benchmark model #39

Open LordRudolf opened 4 years ago

LordRudolf commented 4 years ago

Currently, the bayesian optimization uses gaussian process as the surrogate model. The research shows that random forest underlying process ahcieves fastest convergence: http://ceur-ws.org/Vol-1201/paper-06.pdf