Bayesian Estimation of Causal Effects Using Proxies of a Latent Interference Network

2025-05-13

AI Summary

This paper addresses causal effect estimation under network interference, tackling the practical challenge that the true interference network is unobserved and only noisy, multi-source, or multi-layer proxy networks are available. We propose the first Bayesian causal inference framework explicitly designed for latent interference networks, formalized via a structural causal model that rigorously defines causal effects under population-level interventions. To handle the high-dimensional discrete-continuous mixed posterior distribution, we develop a block Gibbs sampler incorporating local information proposals. Crucially, our method integrates uncertainty from proxy networks—whether corrupted by noise, derived from heterogeneous sources, or structured across multiple layers—directly into the inference process, avoiding biases inherent in conventional approaches that either misuse observed proxies as ground truth or ignore network uncertainty altogether. Numerical experiments demonstrate that our method robustly recovers causal effects with high accuracy even when proxy networks substantially deviate from the true interference structure, significantly outperforming existing baselines.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Network interference occurs when treatments assigned to some units affect the outcomes of others. Traditional approaches often assume that the observed network correctly specifies the interference structure. However, in practice, researchers frequently only have access to proxy measurements of the interference network due to limitations in data collection or potential mismatches between measured networks and actual interference pathways. In this paper, we introduce a framework for estimating causal effects when only proxy networks are available. Our approach leverages a structural causal model that accommodates diverse proxy types, including noisy measurements, multiple data sources, and multilayer networks, and defines causal effects as interventions on population-level treatments. Since the true interference network is latent, estimation poses significant challenges. To overcome them, we develop a Bayesian inference framework. We propose a Block Gibbs sampler with Locally Informed Proposals to update the latent network, thereby efficiently exploring the high-dimensional posterior space composed of both discrete and continuous parameters. We illustrate the performance of our method through numerical experiments, demonstrating its accuracy in recovering causal effects even when only proxies of the interference network are available.

Problem

• Estimating causal effects with proxy network data
• Handling latent interference network in causal inference
• Bayesian framework for high-dimensional posterior exploration

Innovation

• Bayesian framework estimates causal effects via proxies
• Block Gibbs sampler updates latent interference network
• Handles diverse proxy types like noisy multilayer networks