Learning and interpreting asymmetry-labeled DAGs: a case study on COVID-19 fear
This work addresses the limitation of symmetric assumptions in Bayesian networks for researchers in causal inference and data analysis, though it appears incremental as it builds on recently proposed asymmetry-labeled DAGs.
The authors tackled the problem of learning and interpreting asymmetry-labeled DAGs, which extend Bayesian networks by relaxing symmetric conditional independence assumptions, and introduced efficient structural learning algorithms that demonstrated practical utility in a COVID-19 fear dataset.
Bayesian networks are widely used to learn and reason about the dependence structure of discrete variables. However, they are only capable of formally encoding symmetric conditional independence, which in practice is often too strict to hold. Asymmetry-labeled DAGs have been recently proposed to both extend the class of Bayesian networks by relaxing the symmetric assumption of independence and denote the type of dependence existing between the variables of interest. Here, we introduce novel structural learning algorithms for this class of models which, whilst being efficient, allow for a straightforward interpretation of the underlying dependence structure. A comprehensive computational study highlights the efficiency of the algorithms. A real-world data application using data from the Fear of COVID-19 Scale collected in Italy showcases their use in practice.