Masked Gradient-Based Causal Structure Learning

TL;DR

This paper introduces a gradient-based method for learning causal structures from observational data by reformulating SEMs with additive noise, enabling efficient optimization and improved accuracy.

Contribution

It proposes a novel reformulation of SEMs that allows gradient-based learning of causal graphs, leveraging smooth acyclicity constraints and Gumbel-Softmax approximation.

Findings

Effective in synthetic and real datasets

Edges can be thresholded from near-binary entries

Achieves improved performance over existing methods

Abstract

This paper studies the problem of learning causal structures from observational data. We reformulate the Structural Equation Model (SEM) with additive noises in a form parameterized by binary graph adjacency matrix and show that, if the original SEM is identifiable, then the binary adjacency matrix can be identified up to super-graphs of the true causal graph under mild conditions. We then utilize the reformulated SEM to develop a causal structure learning method that can be efficiently trained using gradient-based optimization, by leveraging a smooth characterization on acyclicity and the Gumbel-Softmax approach to approximate the binary adjacency matrix. It is found that the obtained entries are typically near zero or one and can be easily thresholded to identify the edges. We conduct experiments on synthetic and real datasets to validate the effectiveness of the proposed method, and show that it readily includes different smooth model functions and achieves a much improved performance on most datasets considered.