arXiv：cs.LG· Matteo Ceriscioli, Shohei Shimizu, Mallika Gupta, Karthika Mohan·· 6 小时前
因果发现的可识别性结果如何从完整数据推广到缺失数据场景
Generalizing Identifiability Results in Causal Discovery from Complete Data to Missing Data Settings
AI 导读
研究给出因果图在多种缺失机制（含 Missing Not At Random）下仍可识别的通用充分条件：缺失机制之间无因果交互、变量不直接导致自身缺失，且缺失机制满足 faithfulness 或服从广义线性模型。基于此提出两种渐近正确的 Additive Noise Models（ANMs）算法，可恢复潜在因果图并给出相合参数估计。
正文
Abstract:While causal discovery methods have matured for complete data, such algorithms for handling missing values remain limited. We establish general sufficient conditions under which causal graphs remain identifiable (i.e., uniquely determined from the available data distribution) under various missingness categories, including Missing Not At Random, and provide a framework to extend correctness guarantees from causal discovery algorithms for complete data to missing data settings. Our results show that identifiability of causal graphs with no unobserved confounders is preserved when (1) missingness mechanisms do not causally interact, (2) no variable directly causes its own missingness, and (3) either faithfulness holds for missingness mechanisms, or they follow generalized linear models. Guided by these theoretical results, we introduce two asymptotically correct algorithms for Additive Noise Models (ANMs) that recover the underlying causal graph and provide consistent parameter estimates.
| Subjects: | Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Methodology (stat.ME) |
| Cite as: | arXiv:2610.19211 [stat.ML] |
| (or arXiv:2610.19211v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19211 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Matteo Ceriscioli [view email]
[v1]
Tue, 13 Oct 2026 21:24:01 UTC (703 KB)
来源：arXiv：cs.LG · arxiv.org