arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.07575cs.AImath.STstat.MEstat.TH
收藏

基于约束的因果学习条件的一般框架

A General Framework on Conditions for Constraint-based Causal Learning

University College London(伦敦大学学院);Institution Name(机构名称)

Kai Z. Teh, Kayvan Sadeghi, Terry Soo

AI总结:

本文提出一个一般框架,用于研究约束因果学习算法的正确性条件,通过属性概念表示算法,并为PC算法提供精确条件,同时探讨其他因果发现算法的条件及改进方向。

摘要:

大多数基于约束的因果学习算法在特定正确性条件下(如忠实性)能正确返回因果图。通过将任何此类算法表示为属性的概念,本文提供了一个一般框架,用于获得和研究这些算法的正确性条件。从该框架中,我们为PC算法提供了精确的正确性条件,并将其与一些其他现有因果发现算法的正确性条件相关联。该框架还提出了一个设计因果学习算法的范式,允许在设计实际算法之前控制算法的正确性条件。我们证明,最稀疏的马尔可夫表示条件是对于输出祖先图或满足任何现有最小性概念的有向无环图的最弱正确性条件。我们还推论,Pearl最小性对于有意义的因果学习是必要的,但不足以放松忠实性条件,因此需要加强,例如通过纳入背景知识,以实现超越忠实性的因果学习。

英文摘要:

Most constraint-based causal learning algorithms provably return the correct causal graph under certain correctness conditions, such as faithfulness. By representing any constraint-based causal learning algorithm using the notion of a property, we provide a general framework to obtain and study correctness conditions for these algorithms. From the framework, we provide exact correctness conditions for the PC algorithm, which are then related to the correctness conditions of some other existing causal discovery algorithms. The framework also suggests a paradigm for designing causal learning algorithms which allows for the correctness conditions of algorithms to be controlled for before designing the actual algorithm, and has the following implications. We show that the sparsest Markov representation condition is the weakest correctness condition for algorithms that output ancestral graphs or directed acyclic graphs satisfying any existing notions of minimality. We also reason that Pearl-minimality is necessary for meaningful causal learning but not sufficient to relax the faithfulness condition and, as such, has to be strengthened, such as by including background knowledge, for causal learning beyond faithfulness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑