Hybrid Local Causal Discovery

TL;DR

The paper introduces HLCD, a hybrid algorithm for local causal discovery that combines constraint-based and score-based methods to improve accuracy and robustness in identifying local causal structures.

Contribution

HLCD innovatively integrates constraint-based and score-based approaches, effectively addressing cascading errors and local equivalence class issues in local causal discovery.

Findings

HLCD outperforms seven state-of-the-art algorithms on benchmark datasets.

HLCD achieves higher accuracy in identifying local causal structures.

Experimental results demonstrate the effectiveness of the hybrid approach.

Abstract

Local causal discovery aims to learn and distinguish the direct causes and effects of a target variable from observed data. Existing constraint-based local causal discovery methods use AND or OR rules in constructing the local causal skeleton, but using either rule alone is prone to produce cascading errors in the learned local causal skeleton, and thus impacting the inference of local causal relationships. On the other hand, directly applying score-based global causal discovery methods to local causal discovery may randomly return incorrect results due to the existence of local equivalence classes. To address the above issues, we propose a Hybrid Local Causal Discovery algorithm, called HLCD. Specifically, HLCD initially utilizes a constraint-based approach combined with the OR rule to obtain a candidate skeleton and then employs a score-based method to eliminate redundant portions in the candidate skeleton. Furthermore, during the local causal orientation phase, HLCD distinguishes between V-structures and equivalence classes by comparing the local structure scores between the two, thereby avoiding orientation interference caused by local equivalence classes. We conducted extensive experiments with seven state-of-the-art competitors on 14 benchmark Bayesian network datasets, and the experimental results demonstrate that HLCD significantly outperforms existing local causal discovery algorithms.