Webinar: Moritz Schauer (Chalmers/GU) October 20

The first webinar of the fall will be given by Moritz Schauer (Chalmers/GU).

When: Friday October 20, 2023, 11.00 – 11.45

Where: Zoom https://chalmers.zoom.us/j/68608789814 (password: 197131)

Titel: Causal structure learning and sampling using Markov Monte Carlo with momentum.

Abstract: In the context of inferring a Bayesian network structure from observational data, that is inferring a directed acyclic graph (DAG), we devise a non-reversible continuous-time Markov chain 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 and a Markov equivalent likelihood. Joint work with Marcel Wienöbst (Universität zu Lübeck).