policytree: Policy learning via doubly robust empirical welfare maximization over trees

policytree: Policy learning via doubly robust empirical welfare maximization over trees

By
Erik Sverdrup, Ayush Kanodia, Zhengyuan Zhou, Susan Athey
The Journal of Open Source Software. June
20, 2020, Vol. 5, Issue 50, Pages 2243

The problem of learning treatment assignment policies from randomized or observational data arises in many fields. For example, in personalized medicine, we seek to map patient observables (like age, gender, heart pressure, etc.) to a treatment choice using a data-driven rule.