TL;DR本文提出了一种基于有限数据进行稳健因果发现的算法 $k$-PC,通过对两个因果图的条件独立性约束进行比较建立了 $k$-Markov 等价。实验表明,$k$-PC 算法相较于传统 PC 算法,可以在小样本环境中实现更强大的稳健性。
Abstract
Constraint-based causal discovery algorithms learn part of the causal graph
structure by systematically testing conditional independences observed in the
data. These algorithms, such as the PC algorithm and its variants, rely on
graphical characterizations of the so-called equivalence