Accelerating Recursive Partition-Based Causal Structure Learning

TL;DR

This paper introduces a generic refinement strategy for recursive causal discovery algorithms that significantly reduces the number of CI-tests needed, thereby accelerating causal structure learning in large, complex datasets.

Contribution

It proposes a novel refinement method that improves efficiency of recursive causal discovery algorithms, with proven correctness and empirical validation.

Findings

Reduces CI-tests in causal structure refinement

Speeds up causal discovery in large datasets

Maintains high solution quality

Abstract

Causal structure discovery from observational data is fundamental to the causal understanding of autonomous systems such as medical decision support systems, advertising campaigns and self-driving cars. This is essential to solve well-known causal decision making and prediction problems associated with those real-world applications. Recently, recursive causal discovery algorithms have gained particular attention among the research community due to their ability to provide good results by using Conditional Independent (CI) tests in smaller sub-problems. However, each of such algorithms needs a refinement function to remove undesired causal relations of the discovered graphs. Notably, with the increase of the problem size, the computation cost (i.e., the number of CI-tests) of the refinement function makes an algorithm expensive to deploy in practice. This paper proposes a generic causal structure refinement strategy that can locate the undesired relations with a small number of CI-tests, thus speeding up the algorithm for large and complex problems. We theoretically prove the correctness of our algorithm. We then empirically evaluate its performance against the state-of-the-art algorithms in terms of solution quality and completion time in synthetic and real datasets.