An efficient search-and-score algorithm for ancestral graphs using multivariate information scores

Nikita Lagrange, Herve Isambert

Keywords: causal discovery, search-and-score structure learning, latent variable, multivariate information

TL;DR: We propose an efficient search-and-score algorithm for ancestral graphs based multivariate information scores

Abstract: We propose a greedy search-and-score algorithm for ancestral graphs, which include directed as well as bidirected edges, originating from unobserved latent variables. The normalized likelihood score of directed mixed graphs is estimated in terms of multivariate information over relevant subsets of variables, ${C}$, that are connected through collider paths confined to the ancestor set of ${C}$. For computational efficiency, the proposed two-step algorithm relies on local information scores limited to the close surrounding variables of each node (step 1) and edge (step 2). This computational strategy is shown to outperform state-of-the-art causal discovery methods on challenging benchmark datasets.

Primary Area: Causal inference

Submission Number: 19759