跳到正文
arXiv cs.AI· David S. Watson, Sander Beckers·· 4 小时前AI 评分27

非确定性因果模型中的因果概率研究

Probabilities of Causation in Nondeterministic Causal Models

AI 导读

一篇 arXiv 论文研究非确定性因果模型中的因果概率——即固定所有潜在变量后机制仍具有内在随机性的情形。论文针对二元处理与结果,提出通用识别方法:将观察与干预数据表示为潜在随机机制类型的矩约束,通过约束优化计算反事实量的最小/最大值,并推导出若干假设下的解析紧界。案例研究与半合成数据实验表明,所得识别区域往往比确定性模型下更有信息量,并推广了经典因果界。

正文 · 原文

View PDF HTML (experimental)

Abstract:We study probabilities of causation in nondeterministic causal models, where mechanisms may remain intrinsically stochastic even after all latent variables are fixed. Focusing on binary treatments and outcomes, we develop a generic identification program that represents observational and interventional data as moment constraints on latent stochastic mechanism types and computes minimal/maximal values of counterfactual quantities via constrained optimization. Within this framework, we derive sharp analytic bounds on the probabilities of causation under several assumptions tailored to nondeterministic models. Resulting identification regions are often more informative than those attainable under deterministic alternatives. We illustrate the practical relevance of these results in a case study on personalized treatment decisions and experiments on semi-synthetic datasets. Our results clarify how classical bounds relate to nondeterministic causal semantics, generalizing previous works and providing a unified route from qualitative structural restrictions to computable partial identification regions.
Comments: 24 pages, 6 figures
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.18245 [cs.AI]
  (or arXiv:2611.18245v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.18245

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: David Watson [view email]
[v1] Fri, 13 Nov 2026 14:41:49 UTC (835 KB)

来源:arXiv cs.AI · arxiv.org