Causal structure learning with momentum: Sampling distributions over Markov Equivalence Classes
Paper in proceeding, 2024

In the context of inferring a Bayesian network structure (directed acyclic graph, DAG for short), we devise a non-reversible continuous time Markov chain, the “Causal Zig-Zag sampler”, that targets a probability distribution over classes of observationally equivalent (Markov equivalent) DAGs. The classes are represented as completed partially directed acyclic graphs (CPDAGs). The non-reversible Markov chain relies on the operators used in Chickering’s Greedy Equivalence Search (GES) and is endowed with a momentum variable, which improves mixing significantly as we show empirically. The possible target distributions include posterior distributions based on a prior over DAGs and a Markov equivalent likelihood. We offer an efficient implementation wherein we develop new algorithms for listing, counting, uniformly sampling, and applying possible moves of the GES operators, all of which significantly improve upon the state-of-the-art run-time.

DAGs

Causal Structure Learning

MCMC

Causal Discovery

Markov Equivalence Classes

Author

Moritz Schauer

Chalmers, Mathematical Sciences, Applied Mathematics and Statistics

Marcel Wienöbst

Universitaet Zu Lübeck

Proceedings of Machine Learning Research

26403498 (eISSN)

Vol. 246 382-400

The 12th International Conference on Probabilistic Graphical Models
Nijmegen, Netherlands,

Areas of Advance

Information and Communication Technology

Roots

Basic sciences

Subject Categories

Probability Theory and Statistics

More information

Latest update

12/5/2024