Interpretable Off-Policy Learning via Hyperbox Search

TL;DR

This paper introduces an interpretable off-policy learning algorithm using hyperbox search, enabling transparent treatment policies that approximate any function and outperform existing methods in regret, with positive clinical expert feedback.

Contribution

It presents a novel hyperbox search algorithm for interpretable policy learning that guarantees universal approximation and improves regret over state-of-the-art methods.

Findings

Outperforms existing interpretable methods in regret on simulations.

Policies are rated highly interpretable by clinical experts.

The policy class can approximate any measurable function arbitrarily well.

Abstract

Personalized treatment decisions have become an integral part of modern medicine. Thereby, the aim is to make treatment decisions based on individual patient characteristics. Numerous methods have been developed for learning such policies from observational data that achieve the best outcome across a certain policy class. Yet these methods are rarely interpretable. However, interpretability is often a prerequisite for policy learning in clinical practice. In this paper, we propose an algorithm for interpretable off-policy learning via hyperbox search. In particular, our policies can be represented in disjunctive normal form (i.e., OR-of-ANDs) and are thus intelligible. We prove a universal approximation theorem that shows that our policy class is flexible enough to approximate any measurable function arbitrarily well. For optimization, we develop a tailored column generation procedure within a branch-and-bound framework. Using a simulation study, we demonstrate that our algorithm outperforms state-of-the-art methods from interpretable off-policy learning in terms of regret. Using real-word clinical data, we perform a user study with actual clinical experts, who rate our policies as highly interpretable.