Iterative Causal Discovery in the Possible Presence of Latent Confounders and Selection Bias

TL;DR

The paper introduces ICD, a sound and complete iterative algorithm for causal discovery that effectively handles latent confounders and selection bias, requiring fewer tests and improving accuracy over existing methods.

Contribution

The paper presents ICD, a novel iterative causal discovery algorithm that refines causal graphs by increasing conditioning set size, improving efficiency and accuracy in complex scenarios.

Findings

ICD requires fewer conditional independence tests than existing algorithms.

ICD produces more accurate causal graphs in empirical evaluations.

ICD is anytime and guarantees correctness of inferred causal relations.

Abstract

We present a sound and complete algorithm, called iterative causal discovery (ICD), for recovering causal graphs in the presence of latent confounders and selection bias. ICD relies on the causal Markov and faithfulness assumptions and recovers the equivalence class of the underlying causal graph. It starts with a complete graph, and consists of a single iterative stage that gradually refines this graph by identifying conditional independence (CI) between connected nodes. Independence and causal relations entailed after any iteration are correct, rendering ICD anytime. Essentially, we tie the size of the CI conditioning set to its distance on the graph from the tested nodes, and increase this value in the successive iteration. Thus, each iteration refines a graph that was recovered by previous iterations having smaller conditioning sets -- a higher statistical power -- which contributes to stability. We demonstrate empirically that ICD requires significantly fewer CI tests and learns more accurate causal graphs compared to FCI, FCI+, and RFCI algorithms (code is available at https://github.com/IntelLabs/causality-lab).