Optimal Policy Trees

TL;DR

Optimal Policy Trees is a scalable, interpretable method for learning decision trees that prescribe optimal treatments from data, integrating causal inference techniques and achieving state-of-the-art results.

Contribution

The paper introduces a novel approach combining causal inference and optimal decision tree training to learn interpretable, globally optimal prescription policies from data.

Findings

Achieves best-in-class performance on synthetic datasets.

Handles both discrete and continuous treatments effectively.

Provides highly interpretable prescription policies.

Abstract

We propose an approach for learning optimal tree-based prescription policies directly from data, combining methods for counterfactual estimation from the causal inference literature with recent advances in training globally-optimal decision trees. The resulting method, Optimal Policy Trees, yields interpretable prescription policies, is highly scalable, and handles both discrete and continuous treatments. We conduct extensive experiments on both synthetic and real-world datasets and demonstrate that these trees offer best-in-class performance across a wide variety of problems.