Comparative Benchmarking of Causal Discovery Techniques

Abstract

In this paper we present a comprehensive view of prominent causal discovery algorithms, categorized into two main categories (1) assuming acyclic and no latent variables, and (2) allowing both cycles and latent variables, along with experimental results comparing them from three perspectives: (a) structural accuracy, (b) standard predictive accuracy, and (c) accuracy of counterfactual inference. For (b) and (c) we train causal Bayesian networks with structures as predicted by each causal discovery technique to carry out counterfactual or standard predictive inference. We compare causal algorithms on two pub- licly available and one simulated datasets having different sample sizes: small, medium and large. Experiments show that structural accuracy of a technique does not necessarily correlate with higher accuracy of inferencing tasks. Fur- ther, surveyed structure learning algorithms do not perform well in terms of structural accuracy in case of datasets having large number of variables.

arXiv admin note: text overlap with arXiv:1506.07669, arXiv:1611.03977 by other authors

References in corpus (5)

• Being Bayesian about Network Structure
• Causal Inference in the Presence of Latent Variables and Selection Bias
• Searching for Bayesian Network Structures in the Space of Restricted Acyclic Partially Directed Graphs
• Adjacency-Faithfulness and Conservative Causal Inference
• A Review on Algorithms for Constraint-based Causal Discovery