Learning and Testing Exposure Mappings of Interference using Graph Convolutional Autoencoder

TL;DR

This paper introduces a graph convolutional autoencoder to learn data-driven exposure mappings for interference effects in networks and proposes a machine learning-based test to validate these mappings, improving causal inference accuracy.

Contribution

It presents a novel autoencoder approach to learn exposure mappings and a new validity test for these mappings, enhancing the modeling of interference in network experiments.

Findings

The autoencoder effectively captures complex interference patterns.

The validity test accurately assesses exposure mappings in simulations.

Results show improved identification of direct effects under interference.

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.