Statistical Details for the Boosted Tree Platform

This section describes details specific to the Boosted Tree Platform. For statistical details about recursive decision trees, see “Statistical Details for the Partition Platform”.

Statistical Details for the Overfit Penalty

When the response is categorical, a parametric penalty is imposed. For each layer, the estimates minimize the negative log-likelihood plus the penalty value multiplied by the sum of squares of the estimates for each observation. This penalty encourages each new layer not to overfit the training data.