ExMAG: Learning of Maximally Ancestral Graphs
This work addresses causal learning in the presence of confounders, offering an incremental improvement in accuracy and speed for synthetic instances.
The paper tackles the problem of learning maximally ancestral graphs for causal inference with confounders, proposing a score-based branch-and-cut algorithm that achieves higher accuracy and faster runtime on synthetic data compared to state-of-the-art methods.
In mixed graphs, there are both directed and undirected edges. An extension of acyclicity to this mixed-graph setting is known as maximally ancestral graphs. This extension is of considerable interest in causal learning in the presence of confounders. There, directed edges represent a clear direction of causality, while undirected edges represent confounding. We propose a score-based branch-and-cut algorithm for learning maximally ancestral graphs. The algorithm produces more accurate results than state-of-the-art methods, while being faster to run on small and medium-sized synthetic instances.