跳到正文
arXiv cs.AI· Bruno Petrungaro, Anthony C. Constantinou·· 4 小时前AI 评分30

变量时滞时间序列因果发现方法研究

Time series causal discovery with variable lags

AI 导读

arXiv 论文提出一种从时间序列数据学习因果贝叶斯网络结构的方法,可联合学习变量对的时序顺序与主导时滞。方法采用面向混合数据的可分解 BIC 风格评分,结合偏向近期时滞的惩罚项,用带 change-lag 移动的 Tabu 搜索优化,支持缓存与并行邻域评估。

正文 · 原文

View PDF HTML (experimental)

Abstract:Causal Bayesian Networks (CBNs) are a powerful tool for reasoning under uncertainty about complex real-world problems. Such problems evolve over time, responding to external shocks as they occur. To support decision-making, CBNs require a cause-and-effect map of the variables under consideration, known as the network's structure. Learning the graphical structure of a causal model from data remains challenging; learning it from time-series data is even harder because dependencies may arise at different time lags. We construct a method that jointly learns a temporal ordering and a dominant lag for each selected ordered variable pair. Our contribution is a decomposable BIC-style objective score for mixed data that combines node-specific sample sizes with a penalty favouring recent lags. We optimise this objective using a Tabu search augmented with change-lag moves and greedy per-child lag tuning, while its node-wise decomposition supports cached and parallel neighbourhood evaluation. We establish acyclicity and complexity results and derive recovery limits when effects are distributed across multiple lags. In simulations, we show that the proposed method outperforms an inflexible lag-1 score-based learner in most evaluated settings, whereas most specialised methods achieve higher recovery accuracy on data that satisfy their assumptions. On a real-world UK COVID-19 dataset, the method learnt an expected short-lag-skewed structure; nevertheless, 46.8% of selected edges had lags greater than one, showing that longer delays can be retained when their improved fit offsets the penalty. Under its no-contemporaneous-edges and causal-sufficiency assumptions, the method provides a computationally tractable subroutine for edge-specific single-lag optimisation, complementing score-based causal discovery methods.
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2605.04081 [cs.LG]
  (or arXiv:2605.04081v3 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2605.04081

arXiv-issued DOI via DataCite

Submission history

From: Bruno Petrungaro [view email]
[v1] Wed, 15 Apr 2026 09:29:19 UTC (1,410 KB)
[v2] Fri, 8 May 2026 15:18:07 UTC (1,410 KB)
[v3] Fri, 13 Nov 2026 02:26:48 UTC (1,419 KB)

来源:arXiv cs.AI · arxiv.org