Reframed GES with a Neural Conditional Dependence Measure. (arXiv:2206.08531v1 [stat.ML])

Web: http://arxiv.org/abs/2206.08531

In a nonparametric setting, the causal structure is often identifiable only
up to Markov equivalence, and for the purpose of causal inference, it is useful
to learn a graphical representation of the Markov equivalence class (MEC). In
this paper, we revisit the Greedy Equivalence Search (GES) algorithm, which is
widely cited as a score-based algorithm for learning the MEC of the underlying
causal structure. We observe that in order to make the GES algorithm consistent
in a nonparametric setting, it …

arxiv ml neural