Adjustment Identification Distance: A gadjid for Causal Structure Learning

TL;DR

This paper introduces a new framework for causal graph distances, including efficient algorithms and an open-source package, improving scalability and accuracy in evaluating causal discovery algorithms.

Contribution

The paper develops a unified framework for causal graph distances, introduces improved adjustment-based distances, and provides a fast, scalable implementation in gadjid.

Findings

New causal distances with low polynomial time complexity

Open-source package gadjid offers faster evaluation of causal graphs

Framework generalizes existing structural intervention distance

Abstract

Evaluating graphs learned by causal discovery algorithms is difficult: The number of edges that differ between two graphs does not reflect how the graphs differ with respect to the identifying formulas they suggest for causal effects. We introduce a framework for developing causal distances between graphs which includes the structural intervention distance for directed acyclic graphs as a special case. We use this framework to develop improved adjustment-based distances as well as extensions to completed partially directed acyclic graphs and causal orders. We develop new reachability algorithms to compute the distances efficiently and to prove their low polynomial time complexity. In our package gadjid (open source at https://github.com/CausalDisco/gadjid), we provide implementations of our distances; they are orders of magnitude faster with proven lower time complexity than the structural intervention distance and thereby provide a success metric for causal discovery that scales to graph sizes that were previously prohibitive.