arXiv：cs.AI· Arvind Raghavan, Elias Bareinboim·· 1 天前
从反事实数据中进行因果识别：完备性与边界结果
Causal Identification from Counterfactual Data: Completeness and Bounding Results
AI 导读
研究者提出 CTFIDU+ 算法，可从任意一组 Layer 3 反事实分布中识别反事实查询，并证明该算法对此任务具有完备性。基于此，研究确立了从物理可实现分布中可识别反事实的理论极限，并针对无法识别的关键反事实类型，利用可实现反事实数据推导出新的解析边界，模拟验证反事实数据能收紧不可识别量的边界。该成果发表于 ICML'26。
正文
Abstract:Previous work establishing completeness results for counterfactual identification has been circumscribed to the setting where the input data belongs to observational or interventional distributions (Layers 1 and 2 of Pearl's Causal Hierarchy), since it was generally presumed impossible to obtain data from counterfactual distributions, which belong to Layer 3. However, recent work (Raghavan & Bareinboim, 2025) has formally characterized a family of counterfactual distributions which can be directly estimated via experimental methods - a notion they call counterfactual realizability. This leaves open the question of what additional counterfactual quantities now become identifiable, given this new access to (some) Layer 3 data. To answer this question, we develop the CTFIDU+ algorithm for identifying counterfactual queries from an arbitrary set of Layer 3 distributions, and prove that it is complete for this task. Building on this, we establish the theoretical limit of which counterfactuals can be identified from physically realizable distributions, thus implying the fundamental limit to exact causal inference in the non-parametric setting. Finally, given the impossibility of identifying certain critical types of counterfactuals, we derive novel analytic bounds for such quantities using realizable counterfactual data, and corroborate using simulations that counterfactual data helps tighten the bounds for non-identifiable quantities in practice.
| Comments: | published at ICML'26 |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| ACM classes: | F.4.1; G.3 |
| Report number: | Causal AI lab TR-144 |
| Cite as: | arXiv:2602.23541 [cs.AI] |
| (or arXiv:2602.23541v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2602.23541 arXiv-issued DOI via DataCite |
Submission history
From: Arvind Raghavan [view email]
[v1]
Thu, 26 Feb 2026 22:48:12 UTC (1,668 KB)
[v2]
Tue, 3 Mar 2026 20:56:54 UTC (1,668 KB)
[v3]
Fri, 9 Oct 2026 20:21:49 UTC (2,034 KB)
来源：arXiv：cs.AI · arxiv.org