Causal Discovery in a Binary Exclusive-or Skew Acyclic Model: BExSAM

TL;DR

This paper introduces BExSAM, a novel causal discovery method specifically designed for binary data with skewed noise distributions, demonstrating high effectiveness on artificial and real datasets.

Contribution

The paper proposes a new causal model and an efficient approach tailored for binary data with skewed noise, extending causal discovery techniques beyond continuous data.

Findings

High accuracy in artificial data experiments

Effective on real-world binary datasets

Outperforms existing methods for binary causal discovery

Abstract

Discovering causal relations among observed variables in a given data set is a major objective in studies of statistics and artificial intelligence. Recently, some techniques to discover a unique causal model have been explored based on non-Gaussianity of the observed data distribution. However, most of these are limited to continuous data. In this paper, we present a novel causal model for binary data and propose an efficient new approach to deriving the unique causal model governing a given binary data set under skew distributions of external binary noises. Experimental evaluation shows excellent performance for both artificial and real world data sets.