causalml
causalml copied to clipboard
Uplift Forest parameters for multiple treatments
In the section on Multiple Treatments: Splitting Criteria in Rzepakowski 2012 (Decision trees for uplift modeling with single and multiple treatments), the authors describe parameters α, λ, and γ for setting the relative importance of specific treatments when fitting a tree. The causalml
docs don't seem to mention this (though I could have missed it).
Do the Uplift Tree/Forest Classifiers allow users to set the relative importance of multiple treatments? If not, what default values are used for α, λ, and γ?