Reliable Decision Support using Counterfactual Models

TL;DR

This paper introduces counterfactual models, including the Counterfactual Gaussian Process, to improve decision support by accurately predicting outcomes under hypothetical actions, addressing limitations of supervised learning.

Contribution

It proposes a novel counterfactual learning objective and the CGP model for better future outcome prediction in decision-making contexts.

Findings

Counterfactual models outperform supervised models in risk prediction.

CGP effectively predicts future trajectories under different actions.

Improves 'what if?' reasoning for personalized treatment planning.

Abstract

Decision-makers are faced with the challenge of estimating what is likely to happen when they take an action. For instance, if I choose not to treat this patient, are they likely to die? Practitioners commonly use supervised learning algorithms to fit predictive models that help decision-makers reason about likely future outcomes, but we show that this approach is unreliable, and sometimes even dangerous. The key issue is that supervised learning algorithms are highly sensitive to the policy used to choose actions in the training data, which causes the model to capture relationships that do not generalize. We propose using a different learning objective that predicts counterfactuals instead of predicting outcomes under an existing action policy as in supervised learning. To support decision-making in temporal settings, we introduce the Counterfactual Gaussian Process (CGP) to predict the counterfactual future progression of continuous-time trajectories under sequences of future actions. We demonstrate the benefits of the CGP on two important decision-support tasks: risk prediction and "what if?" reasoning for individualized treatment planning.