Mixed Graphical Models for Causal Analysis of Multi-modal Variables

TL;DR

This paper introduces new hybrid algorithms for learning causal graphical models from multi-modal data, effectively handling mixed variable types and outperforming existing methods in speed and accuracy.

Contribution

The paper proposes a novel conditional independence test and hybrid learning strategies specifically designed for mixed data types in causal graphical models.

Findings

Hybrid methods are faster and more accurate than existing directed graph estimation techniques.

The new conditional independence test improves causal discovery in mixed data.

Performance improvements are validated on synthetic datasets.

Abstract

Graphical causal models are an important tool for knowledge discovery because they can represent both the causal relations between variables and the multivariate probability distributions over the data. Once learned, causal graphs can be used for classification, feature selection and hypothesis generation, while revealing the underlying causal network structure and thus allowing for arbitrary likelihood queries over the data. However, current algorithms for learning sparse directed graphs are generally designed to handle only one type of data (continuous-only or discrete-only), which limits their applicability to a large class of multi-modal biological datasets that include mixed type variables. To address this issue, we developed new methods that modify and combine existing methods for finding undirected graphs with methods for finding directed graphs. These hybrid methods are not only faster, but also perform better than the directed graph estimation methods alone for a variety of parameter settings and data set sizes. Here, we describe a new conditional independence test for learning directed graphs over mixed data types and we compare performances of different graph learning strategies on synthetic data.