Flow-based Generative Modeling of Potential Outcomes and Counterfactuals
About
Predicting potential and counterfactual outcomes from observational data is central to individualized decision-making, particularly in clinical settings where treatment choices must be tailored to each patient rather than guided solely by population averages. We propose PO-Flow, a continuous normalizing flow (CNF) framework for causal inference that jointly models potential outcome distributions and factual-conditioned counterfactual outcomes. Trained via flow matching, PO-Flow provides a unified approach to individualized potential outcome prediction, conditional average treatment effect estimation, and counterfactual prediction. By encoding an observed factual outcome and decoding under an alternative treatment, PO-Flow provides an encode-decode mechanism for factual-conditioned counterfactual prediction. In addition, PO-Flow supports likelihood-based evaluation of potential outcomes, enabling uncertainty-aware assessment of predictions. A supporting recovery guarantee is established under certain assumptions, and empirical results on benchmark datasets demonstrate strong performance across a range of causal inference tasks within the potential outcomes framework.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Individual Treatment Effect Estimation | IHDP (out-of-sample) | -- | 45 | |
| CATE estimation | ACIC 2016 (Out-sample) | PEHE0.82 | 22 | |
| CATE estimation | IHDP (In-sample) | PEHE0.41 | 22 | |
| CATE estimation | IHDP (Out-sample) | PEHE0.45 | 22 | |
| Potential Outcome Estimation | ACIC in-sample 2018 | RMSE0.04 | 19 | |
| Potential Outcome Estimation | ACIC out-of-sample 2018 | RMSE0.05 | 19 | |
| Potential Outcome Estimation | IHDP (In-sample) | RMSE0.96 | 19 | |
| Potential Outcome Estimation | IBM in-sample | RMSE0.04 | 19 | |
| Potential Outcome Estimation | IBM out-of-sample | RMSE0.04 | 19 | |
| CATE estimation | ACIC 2016 (In-sample) | -- | 13 |