blackboost.regr

blackboost.regr-class(mlr)

R Documentation

Wrapped learner for Boosting with Regression Trees from package mboost for regression problems.

Common hyperparameters:

mstop: Integer, giving the number of initial boosting iterations.
nu: Double (between 0 and 1), defining the step size or shrinkage parameter.
constraint: Logical, indicating whether the working responses should be restricted to (-1, +1).
risk: Character, indicating how the empirical risk should be computed for each boosting iteration.
center: Logical, indicating if the numerical covariates should be mean centered before fitting.