构建可生成干预数据的时序因果模型，推动时序因果基础模型发展
Interventional Time Series Priors for Causal Foundation Models
- 提出CausalTimePrior框架，生成带观测与干预数据的时序因果模型
- 训练的PFNs在未见模型上实现上下文因果效应估计，准确率超基线
- 支持复杂动态结构，适合时序因果推断研究者使用
先验数据拟合网络（PFNs）已成为表格式因果推断的强大基础模型，但其在时序数据上的扩展受限于缺乏提供干预目标的合成数据生成器。现有时序基准仅生成具有真实因果图的观测数据，缺少训练因果基础模型所需的干预数据。为此，我们提出 extbf{CausalTimePrior}，一种生成配对观测与干预时序数据的原理性框架。该先验支持可配置的因果图结构、非线性自回归机制、状态切换动态及多种干预类型（硬干预、软干预、时变干预）。实验表明，基于CausalTimePrior训练的PFNs可在未见时序因果模型（TSCMs）上实现上下文因果效应估计，为时序因果推断的基础模型提供了可行路径。
原文摘要 · Abstract (English)
Prior-data fitted networks (PFNs) have emerged as powerful foundation models for tabular causal inference, yet their extension to time series remains limited by the absence of synthetic data generators that provide interventional targets. Existing time series benchmarks generate observational data with ground-truth causal graphs but lack the interventional data required for training causal foundation models. To address this, we propose \textbf{CausalTimePrior}, a principled framework for generating synthetic temporal structural causal models (TSCMs) with paired observational and interventional time series. Our prior supports configurable causal graph structures, nonlinear autoregressive mechanisms, regime-switching dynamics, and multiple intervention types (hard, soft, time-varying). We demonstrate that PFNs trained on CausalTimePrior can perform in-context causal effect estimation on held-out TSCMs, establishing a pathway toward foundation models for time series causal inference.
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。