A Meta Learning Approach to Discerning Causal Graph Structure

TL;DR

This paper introduces a meta-learning framework that infers causal directions between variables using a stochastic graph model, demonstrating robustness to data scarcity and the ability to handle complex, latent-variable-influenced structures.

Contribution

It presents a novel meta-learning approach with a stochastic graph representation for causal inference, capable of handling complex graphs and latent confounders.

Findings

Robust to modest data scarcity

Effective in inferring causal direction and existence

Ensemble method reduces outcome variability

Abstract

We explore the usage of meta-learning to derive the causal direction between variables by optimizing over a measure of distribution simplicity. We incorporate a stochastic graph representation which includes latent variables and allows for more generalizability and graph structure expression. Our model is able to learn causal direction indicators for complex graph structures despite effects of latent confounders. Further, we explore robustness of our method with respect to violations of our distributional assumptions and data scarcity. Our model is particularly robust to modest data scarcity, but is less robust to distributional changes. By interpreting the model predictions as stochastic events, we propose a simple ensemble method classifier to reduce the outcome variability as an average of biased events. This methodology demonstrates ability to infer the existence as well as the direction of a causal relationship between data distributions.