AI RESEARCH
GDR-learners: Orthogonal Learning of Generative Models for Potential Outcomes
arXiv CS.LG
•
ArXi:2509.22953v2 Announce Type: replace Various deep generative models have been proposed to estimate potential outcomes distributions from observational data. However, none of them have the favorable theoretical property of general Neyman-orthogonality and, associated with it, quasi-oracle efficiency and double robustness. In this paper, we