arXiv cs.AI· Ang Li, Scott Mueller, Xin Shu, Judea Pearl·· 13 小时前AI 评分18
因果量的 ε-可识别性
Epsilon-Identifiability of Causal Quantities
AI 导读
论文提出 ε-可识别性（epsilon-identifiability），用于在数据无法完全识别因果概率时实现近似识别：当特定子群体的行为可被限制在足够窄的范围内时，原本不可识别的因果效应与反事实概率可被 ε-识别，且这类限制条件通常易于测量和合理假设。该方法还被应用于单位选择问题（unit selection problem）。
正文 · 原文
Abstract:Identifying the effects of causes and causes of effects is vital in virtually every scientific field. Often, however, the needed probabilities may not be fully identifiable from the available data sources. This paper shows how approximate identifiability is still possible for several probabilities of causation. We term this epsilon-identifiability and demonstrate its usefulness in cases where the behavior of certain subpopulations can be restricted within sufficiently narrow bounds. In particular, we show how unidentifiable causal effects and counterfactual probabilities can be epsilon-identified when such allowances are made. Often, these allowances are easily measured and reasonably assumed. Finally, epsilon-identifiability is applied to the unit selection problem.
| Subjects: | Artificial Intelligence (cs.AI); Statistics Theory (math.ST) |
| Cite as: | arXiv:2301.12022 [cs.AI] |
| (or arXiv:2301.12022v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2301.12022 arXiv-issued DOI via DataCite |
Submission history
From: Ang Li [view email]
[v1]
Fri, 27 Jan 2023 23:16:57 UTC (32 KB)
[v2]
Tue, 27 Oct 2026 03:24:17 UTC (38 KB)
来源：arXiv cs.AI · arxiv.org