A Graph Autoencoder Approach to Causal Structure Learning

TL;DR

This paper introduces a graph autoencoder-based gradient method for causal structure learning that handles nonlinear models and vector variables, outperforming existing methods especially on large graphs.

Contribution

It generalizes recent gradient-based causal discovery methods into a flexible autoencoder framework suitable for complex data structures.

Findings

Outperforms other gradient-based methods on synthetic datasets.

Achieves near-linear training time on large graphs.

Handles nonlinear structural equations and vector-valued variables.

Abstract

Causal structure learning has been a challenging task in the past decades and several mainstream approaches such as constraint- and score-based methods have been studied with theoretical guarantees. Recently, a new approach has transformed the combinatorial structure learning problem into a continuous one and then solved it using gradient-based optimization methods. Following the recent state-of-the-arts, we propose a new gradient-based method to learn causal structures from observational data. The proposed method generalizes the recent gradient-based methods to a graph autoencoder framework that allows nonlinear structural equation models and is easily applicable to vector-valued variables. We demonstrate that on synthetic datasets, our proposed method outperforms other gradient-based methods significantly, especially on large causal graphs. We further investigate the scalability and efficiency of our method, and observe a near linear training time when scaling up the graph size.