Causal Forest

A Causal Forest is made of several causal trees.

Criterion

Try Mean Square Error Try Heterogeneity

The criterion used to choose the split at each node.

Honest framework

Whether each tree should be trained in an honest manner, i.e. the training set is split into two equal sized subsets, A and B.
All samples in A are used to create the split structure and all samples in B are used to calculate the value of each node in the tree.
Recommended, except for small datasets where more samples help build significantly better splits.

Feature sampling strategy

Adjusts the number of features to sample at each split.

Parallelism

Number of cores used for parallel training. Using more cores leads to faster training but at the expense of more memory consumption, especially for large training datasets.