English

Flow-based Generative Modeling of Potential Outcomes and Counterfactuals

Machine Learning 2026-04-16 v4 Machine Learning

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 datasets demonstrate strong performance across a range of causal inference tasks within the potential outcomes framework.

Keywords

Cite

@article{arxiv.2505.16051,
  title  = {Flow-based Generative Modeling of Potential Outcomes and Counterfactuals},
  author = {Dongze Wu and David I. Inouye and Yao Xie},
  journal= {arXiv preprint arXiv:2505.16051},
  year   = {2026}
}

Comments

Accepted at 2026 IEEE International Symposium on Information Theory (ISIT 2026)