Distilling Heterogeneity: From Explanations of Heterogeneous Treatment Effect Models to Interpretable Policies

11/05/2021
by   Han Wu, et al.
14

Internet companies are increasingly using machine learning models to create personalized policies which assign, for each individual, the best predicted treatment for that individual. They are frequently derived from black-box heterogeneous treatment effect (HTE) models that predict individual-level treatment effects. In this paper, we focus on (1) learning explanations for HTE models; (2) learning interpretable policies that prescribe treatment assignments. We also propose guidance trees, an approach to ensemble multiple interpretable policies without the loss of interpretability. These rule-based interpretable policies are easy to deploy and avoid the need to maintain a HTE model in a production environment.

READ FULL TEXT
research
06/21/2022

Interpretable Deep Causal Learning for Moderation Effects

In this extended abstract paper, we address the problem of interpretabil...
research
09/21/2023

Survival causal rule ensemble method considering the main effect for estimating heterogeneous treatment effects

With an increasing focus on precision medicine in medical research, nume...
research
03/12/2021

Evidence-Based Policy Learning

The past years have seen seen the development and deployment of machine-...
research
07/27/2023

Causal rule ensemble method for estimating heterogeneous treatment effect with consideration of main effects

This study proposes a novel framework based on the RuleFit method to est...
research
06/16/2022

Benchmarking Heterogeneous Treatment Effect Models through the Lens of Interpretability

Estimating personalized effects of treatments is a complex, yet pervasiv...
research
01/24/2019

Learning Interpretable Models with Causal Guarantees

Machine learning has shown much promise in helping improve the quality o...
research
03/04/2022

Interpretable Off-Policy Learning via Hyperbox Search

Personalized treatment decisions have become an integral part of modern ...

Please sign up or login with your details

Forgot password? Click here to reset