通过可观察神经微分方程实现连续时间因果预测，解决隐藏混杂偏倚问题。
Observable Neural ODEs for Identifiable Causal Forecasting in Continuous Time
- 构建可观察的神经微分方程模型，确保潜变量可从观测数据重构。
- 在癌症、脓毒症等数据上，对不同治疗路径的预测效果优于现有序列模型。
- 适用于医疗决策等需连续时间因果推断的场景，尤其关注隐变量可识别性。
连续时间序列决策中的因果推断受隐藏混杂因素影响。我们证明，在具有时变干预的潜变量状态空间模型中，潜变量动态的可观测性是识别动态治疗效应的必要条件，将控制理论中的可观测性与因果可识别性关联起来，即使隐藏混杂因素同时影响处理和结果。我们推导出一种连续时间调整公式，通过测量模型、潜变量动力学和基于观测历史的潜变量滤波分布表达治疗轨迹下的潜在结果分布。我们提出可观测神经微分方程（ObsNODEs），一种在可观测规范形式下的神经微分方程模型，用于因果预测。ObsNODEs 能学习可从观测中重构状态的连续时间动态，从而实现对替代治疗路径的结果预测。在合成癌症数据、基于 MIMIC-IV 的半合成数据以及真实脓毒症数据上的实验表明，其性能显著优于近期序列模型。
原文摘要 · Abstract (English)
Causal inference in continuous-time sequential decision problems is challenged by hidden confounders. We show that, in latent state-space models with time-varying interventions, observability of the latent dynamics from observed data is necessary for identifying dynamic treatment effects, linking control-theoretic observability to causal identifiability, even when hidden confounders affect both treatments and outcomes. We derive a continuous-time adjustment formula expressing potential outcome distributions under treatment trajectories via the measurement model, latent dynamics, and the filtering distribution over latent states given observed histories. We propose Observable Neural ODEs (ObsNODEs), Neural ODE models in observable normal form for causal forecasting. ObsNODEs learn continuous-time dynamics with states reconstructible from observations, enabling outcome prediction under alternative treatment paths. Experiments on synthetic cancer data, semi-synthetic data based on MIMIC-IV, and real-world sepsis data show strong performance over recent sequence models.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。