Reliable Decision Support using Counterfactual Models

2017/03/30 by Peter Schulam, Schulam, Peter, Suchi Saria +1 · 9 citations
Decision Sciences · #Artificial Intelligence (cs.AI) #Complex Systems and Decision Making #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.1703.10651

openalex publication_date 2017/03/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Cited by

• The Impact of Time Series Length and Discretization on Longitudinal\n Causal Estimation Methods
• Scaling Gaussian Processes with Derivative Information Using Variational Inference
• Cause-Effect Deep Information Bottleneck For Systematically Missing Covariates
• Calibrating Healthcare AI: Towards Reliable and Interpretable Deep Predictive Models
• Risk Variance Penalization
• Errors-in-variables Modeling of Personalized Treatment-Response Trajectories
• When the Oracle Misleads: Modeling the Consequences of Using Observable Rather than Potential Outcomes in Risk Assessment Instruments
• I-SPEC: An End-to-End Framework for Learning Transportable, Shift-Stable Models
• Active Learning for Decision-Making from Imbalanced Observational Data

Related

• Clinical outcome prediction under hypothetical interventions -- a representation learning framework for counterfactual reasoning
• Counterfactual Explanations in Sequential Decision Making Under\n Uncertainty
• Adversary-Free Counterfactual Prediction via Information-Regularized Representations
• Estimating and evaluating counterfactual prediction models
• Counterfactual Reasoning for Fair Clinical Risk Prediction