Causal structure learning with momentum: Sampling distributions over Markov Equivalence Classes
Paper i 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

Markov Equivalence Classes

Causal Structure Learning

Causal Discovery

MCMC

Författare

Moritz Schauer

Chalmers, Matematiska vetenskaper, Tillämpad matematik och statistik

Marcel Wienöbst

Universitaet Zu Lübeck

Proceedings of Machine Learning Research

26403498 (eISSN)

246 382-400

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

Styrkeområden

Informations- och kommunikationsteknik

Fundament

Grundläggande vetenskaper

Ämneskategorier

Sannolikhetsteori och statistik

Mer information

Skapat

2024-09-28