[2505.16051] Flow-based Generative Modeling of Potential Outcomes and Counterfactuals

Statistics > Machine Learning arXiv:2505.16051 (stat) [Submitted on 21 May 2025 (v1), last revised 15 Apr 2026 (this version, v4)] Title:Flow-based Generative Modeling of Potential Outcomes and Counterfactuals Authors:Dongze Wu, David I. Inouye, Yao Xie View a PDF of the paper titled Flow-based Generative Modeling of Potential Outcomes and Counterfactuals, by Dongze Wu and 2 other authors View PDF HTML (experimental) Abstract: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 data...