aiqbench analysis
This paper introduces Observable Neural ODEs (ObsNODEs), a novel framework that combines continuous-time dynamics with causal inference to handle hidden confounding in sequential decision problems. By leveraging observability of latent states, the method enables robust forecasting under alternative treatment paths, validated across synthetic and clinical datasets. This work represents a significant advancement in applying neural ordinary differential equations to solve complex causal identification challenges in dynamic systems.
Observable Neural ODEs for Identifiable Causal Forecasting in Continuous Time
arXiv:2604.26070v3 Announce Type: replace Abstract: Causal inference in continuous-time sequential decision problems is challenged by hidden confounding and partially observed states. We show that, under explicit structural assumptions, observability of the latent state enables identification of dynamic treatment effects through a continuous-time conditional front-door adjustment, even in the presence of hidden confounding. We derive a general adjustment formula and show that it reduces to a tractable state-space formula when unobserved contemporaneous disturbances are temporally uncorrelated. This formula expresses potential-outcome distributions under alternative treatment trajectories through the measurement model, latent dynamics, and the filtering distribution over latent states. We propose Observable Neural ODEs (ObsNODEs), Neural ODE models in observable normal form that implement this tractable adjustment for causal forecasting. ObsNODEs learn continuous-time dynamics with states reconstructible from observations, enabling outcome prediction under alternative treatment paths. Experiments on synthetic, semi-synthetic, and real-world clinical data demonstrate strong performance over recent sequence models, including external validation.
Abstract:Causal inference in continuous-time sequential decision problems is challenged by hidden confounding and partially observed states. We show that, under explicit structural assumptions, observability of the latent state enables identification of dynamic treatment effects through a continuous-time conditional front-door adjustment, even in the presence of hidden confounding. We derive a general adjustment formula and show that it reduces to a tractable state-space formula when unobserved contemporaneous disturbances are temporally uncorrelated. This formula expresses potential-outcome distributions under alternative treatment trajectories through the measurement model, latent dynamics, and the filtering distribution over latent states. We propose Observable Neural ODEs (ObsNODEs), Neural ODE models in observable normal form that implement this tractable adjustment for causal forecasting. ObsNODEs learn continuous-time dynamics with states reconstructible from observations, enabling outcome prediction under alternative treatment paths. Experiments on synthetic, semi-synthetic, and real-world clinical data demonstrate strong performance over recent sequence models, including external validation.
Machine Learning (cs.LG); Optimization and Control (math.OC); Statistics Theory (math.ST); Quantitative Methods (q-bio.QM)
arXiv:2604.26070 [cs.LG]
(or arXiv:2604.26070v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2604.26070 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 249 words
Related coverage
Identifiability Analysis of Linear ODE Systems with Hidden Confounders
arXiv · Oct 7, 2026
Dharma vs Adharma - The Perplexity of Being Alive
hillpost.in · Aug 25, 2026
The Entangled Dimensions of Decision Theory
LessWrong · Jul 30, 2026
Nonlocal Bayesian Modeling of Continuous Spatio-Temporal Dynamics
arXiv · Jun 15, 2026