Learning and Testing Exposure Mappings of Interference using Graph Convolutional Autoencoder

2026-01-09

AI Summary

This study addresses the limitations of traditional approaches that rely on prespecified, simplistic exposure mappings, which often fail to capture complex interference effects in networked settings. To overcome this challenge, the authors introduce graph neural networks into the causal inference framework and propose a data-driven method for learning and validating exposure mappings. Specifically, they employ a graph convolutional autoencoder to infer individual exposure statuses directly from observational data and integrate conditional independence tests to assess the identification validity of user-defined exposure mappings. This integrated approach enables more accurate estimation of direct treatment effects. Simulation experiments demonstrate that the proposed method achieves strong statistical power and robustness even with limited sample sizes.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Interference or spillover effects arise when an individual's outcome (e.g., health) is influenced not only by their own treatment (e.g., vaccination) but also by the treatment of others, creating challenges for evaluating treatment effects. Exposure mappings provide a framework to study such interference by explicitly modeling how the treatment statuses of contacts within an individual's network affect their outcome. Most existing research relies on a priori exposure mappings of limited complexity, which may fail to capture the full range of interference effects. In contrast, this study applies a graph convolutional autoencoder to learn exposure mappings in a data-driven way, which exploit dependencies and relations within a network to more accurately capture interference effects. As our main contribution, we introduce a machine learning-based test for the validity of exposure mappings and thus test the identification of the direct effect. In this testing approach, the learned exposure mapping is used as an instrument to test the validity of a simple, user-defined exposure mapping. The test leverages the fact that, if the user-defined exposure mapping is valid (so that all interference operates through it), then the learned exposure mapping is statistically independent of any individual's outcome, conditional on the user-defined exposure mapping. We assess the finite-sample performance of this proposed validity test through a simulation study.

Problem

• interference
• exposure mapping
• treatment effect
• spillover effects
• causal inference

Innovation

• graph convolutional autoencoder
• exposure mapping
• interference
• spillover effects
• causal inference