Multi-granularity Causal Structure Learning

TL;DR

This paper introduces MgCSL, a novel multi-granularity causal learning method that effectively models complex causal relationships in high-dimensional data, demonstrated on fMRI datasets with superior performance.

Contribution

The paper proposes MgCSL, combining sparse auto-encoders and multilayer perceptrons to uncover causal structures across multiple granularities, addressing high-dimensionality challenges.

Findings

MgCSL outperforms baseline methods in causal discovery accuracy.

It effectively uncovers explainable causal links in fMRI data.

The method reduces computational complexity with a simplified acyclicity constraint.

Abstract

Unveil, model, and comprehend the causal mechanisms underpinning natural phenomena stand as fundamental endeavors across myriad scientific disciplines. Meanwhile, new knowledge emerges when discovering causal relationships from data. Existing causal learning algorithms predominantly focus on the isolated effects of variables, overlook the intricate interplay of multiple variables and their collective behavioral patterns. Furthermore, the ubiquity of high-dimensional data exacts a substantial temporal cost for causal algorithms. In this paper, we develop a novel method called MgCSL (Multi-granularity Causal Structure Learning), which first leverages sparse auto-encoder to explore coarse-graining strategies and causal abstractions from micro-variables to macro-ones. MgCSL then takes multi-granularity variables as inputs to train multilayer perceptrons and to delve the causality between variables. To enhance the efficacy on high-dimensional data, MgCSL introduces a simplified acyclicity constraint to adeptly search the directed acyclic graph among variables. Experimental results show that MgCSL outperforms competitive baselines, and finds out explainable causal connections on fMRI datasets.