Using Embeddings to Correct for Unobserved Confounding in Networks

TL;DR

This paper introduces a method that leverages network embeddings as proxies for unobserved confounders to improve causal inference in network data, validated through experiments on social network datasets.

Contribution

It proposes a novel approach that reduces causal inference with unobserved confounders to semi-supervised prediction using network embeddings, enabling valid causal estimates.

Findings

Method yields valid inferences under weak conditions.

Network embeddings effectively serve as proxies for unobserved confounders.

Validated on semi-synthetic social network data.

Abstract

We consider causal inference in the presence of unobserved confounding. We study the case where a proxy is available for the unobserved confounding in the form of a network connecting the units. For example, the link structure of a social network carries information about its members. We show how to effectively use the proxy to do causal inference. The main idea is to reduce the causal estimation problem to a semi-supervised prediction of both the treatments and outcomes. Networks admit high-quality embedding models that can be used for this semi-supervised prediction. We show that the method yields valid inferences under suitable (weak) conditions on the quality of the predictive model. We validate the method with experiments on a semi-synthetic social network dataset. Code is available at github.com/vveitch/causal-network-embeddings.