一次完成高维事件序列的多标签因果发现，速度快且可解释。
One-Shot Multi-Label Causal Discovery in High-Dimensional Event Sequences
- 用两个预训练Transformer做密度估计，一次推断每条序列的马尔可夫边界
- 在474个标签、29,100个事件上，分钟级完成因果结构推断
- 适合需要快速诊断的生产级系统，如汽车故障分析
理解包含数千种稀疏事件类型的事件序列中的因果关系，在医疗、网络安全和车辆诊断等领域至关重要，但现有方法难以扩展。我们提出OSCAR，一种一次完成的因果自回归方法，利用两个预训练Transformer作为密度估计器，推断每条序列的马尔可夫边界。该方法无需昂贵的全局条件独立测试，实现高效并行因果发现。在包含474个标签、29,100个事件的真实汽车数据集上，OSCAR可在几分钟内恢复可解释的因果结构，而传统方法无法扩展，使生产级科学诊断成为可能。
原文摘要 · Abstract (English)
Understanding causality in event sequences with thousands of sparse event types is critical in domains such as healthcare, cybersecurity, or vehicle diagnostics, yet current methods fail to scale. We present OSCAR, a one-shot causal autoregressive method that infers per-sequence Markov Boundaries using two pretrained Transformers as density estimators. This enables efficient, parallel causal discovery without costly global CI testing. On a real-world automotive dataset with 29,100 events and 474 labels, OSCAR recovers interpretable causal structures in minutes, while classical methods fail to scale, enabling practical scientific diagnostics at production scale.
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。