Graph Infomax Adversarial Learning for Treatment Effect Estimation with Networked Observational Data

TL;DR

This paper introduces GIAL, a novel adversarial learning model that leverages network structure to improve treatment effect estimation from observational data, effectively addressing hidden confounders and network imbalance.

Contribution

The paper proposes a new Graph Infomax Adversarial Learning model specifically designed for treatment effect estimation using networked observational data, accounting for network imbalance.

Findings

GIAL outperforms existing methods on benchmark datasets.

The model effectively captures hidden confounders.

Network imbalance is addressed successfully.

Abstract

Treatment effect estimation from observational data is a critical research topic across many domains. The foremost challenge in treatment effect estimation is how to capture hidden confounders. Recently, the growing availability of networked observational data offers a new opportunity to deal with the issue of hidden confounders. Unlike networked data in traditional graph learning tasks, such as node classification and link detection, the networked data under the causal inference problem has its particularity, i.e., imbalanced network structure. In this paper, we propose a Graph Infomax Adversarial Learning (GIAL) model for treatment effect estimation, which makes full use of the network structure to capture more information by recognizing the imbalance in network structure. We evaluate the performance of our GIAL model on two benchmark datasets, and the results demonstrate superiority over the state-of-the-art methods.