Causal Structure Learning

TL;DR

This paper reviews methods for learning causal structures from data, emphasizing their assumptions and empirical performance, to improve understanding and decision-making in complex systems.

Contribution

It provides a comparative analysis of recent causal structure learning algorithms and discusses their underlying assumptions and empirical effectiveness.

Findings

Different algorithms have varying assumptions affecting their applicability.

Empirical performance varies across scenarios and methods.

Understanding assumptions is crucial for selecting appropriate causal models.

Abstract

Graphical models can represent a multivariate distribution in a convenient and accessible form as a graph. Causal models can be viewed as a special class of graphical models that not only represent the distribution of the observed system but also the distributions under external interventions. They hence enable predictions under hypothetical interventions, which is important for decision making. The challenging task of learning causal models from data always relies on some underlying assumptions. We discuss several recently proposed structure learning algorithms and their assumptions, and compare their empirical performance under various scenarios.