OrthoGen: A Generative Orthogonal Learner for Time-Varying Treatments

2026-10-07

AI Summary

This study addresses the challenge of time-varying confounding bias in estimating conditional distributions of potential outcomes under time-varying treatments. To this end, it proposes a generative recursive g-computation strategy and constructs OrthoGen, an orthogonal and doubly robust generative learner. As the first generative orthogonal learner tailored for time-varying treatments, OrthoGen integrates Neyman orthogonality with flexible generative architectures, such as normalizing flows and diffusion models, to overcome the limitations of conventional mean-based modeling. This integration achieves rate double robustness and quasi-oracle efficiency. Extensive experiments conducted on synthetic, semi-synthetic, and real-world datasets comprehensively validate the superior effectiveness and robustness of the proposed method.

Abstract

Estimating conditional distributional potential outcomes (CDPOs) over time is important in medicine (e.g., to estimate patient-specific risks under different treatment sequences). However, this task is challenging because of time-varying confounding, yet existing adjustment strategies for this task are limited. In this paper, we aim to learn CDPOs under time-varying treatments using flexible generative models. Our contributions are two-fold. (1) We introduce a tailored adjustment strategy for our setting, namely, generative recursive g-computation. Our adjustment strategy recursively propagates full conditional outcome distributions rather than conditional means, modeling the variables of interest directly rather than full trajectories. Building on our adjustment strategy, we formulate simple generative learners for CDPO estimation. However, these learners can be sensitive to nuisance estimation errors, which motivates an orthogonal learner. (2) We thus introduce OrthoGen, a Neyman-orthogonal and doubly robust generative learner. Importantly, we show that OrthoGen further achieves rate double robustness and quasi-oracle efficiency under suitable conditions. Our learners are flexible and can be instantiated with different generative backbones (e.g., normalizing flows and diffusion models). Across experiments with synthetic, semi-synthetic and real-world datasets, we find that OrthoGen is highly effective. To the best of our knowledge, we are the first to propose a generative orthogonal learner for estimating CDPOs under time-varying treatments.

Problem

• Conditional distributional potential outcomes
• Time-varying treatments
• Time-varying confounding
• Causal inference
• Generative models

Innovation

• Generative Orthogonal Learner
• Conditional Distributional Potential Outcomes
• Time-Varying Treatments
• Recursive G-Computation
• Double Robustness