Graph Disentangle Causal Model: Enhancing Causal Inference in Networked Observational Data

TL;DR

This paper introduces GDC, a novel graph-based causal model that disentangles features into adjustment and confounder representations to improve individual treatment effect estimation in networked observational data.

Contribution

The paper proposes a new framework called GDC that separates features into causal components and employs specialized graph modules for better ITE estimation in network settings.

Findings

GDC outperforms existing methods on two networked datasets.

Disentangling features improves causal inference accuracy.

The model effectively isolates true causal factors from confounders.

Abstract

Estimating individual treatment effects (ITE) from observational data is a critical task across various domains. However, many existing works on ITE estimation overlook the influence of hidden confounders, which remain unobserved at the individual unit level. To address this limitation, researchers have utilized graph neural networks to aggregate neighbors' features to capture the hidden confounders and mitigate confounding bias by minimizing the discrepancy of confounder representations between the treated and control groups. Despite the success of these approaches, practical scenarios often treat all features as confounders and involve substantial differences in feature distributions between the treated and control groups. Confusing the adjustment and confounder and enforcing strict balance on the confounder representations could potentially undermine the effectiveness of outcome prediction. To mitigate this issue, we propose a novel framework called the \textit{Graph Disentangle Causal model} (GDC) to conduct ITE estimation in the network setting. GDC utilizes a causal disentangle module to separate unit features into adjustment and confounder representations. Then we design a graph aggregation module consisting of three distinct graph aggregators to obtain adjustment, confounder, and counterfactual confounder representations. Finally, a causal constraint module is employed to enforce the disentangled representations as true causal factors. The effectiveness of our proposed method is demonstrated by conducting comprehensive experiments on two networked datasets.