Double Machine Learning for Causal Inference under Shared-State Interference

2025-04-10

AI Summary

This paper addresses interference among units in interactive environments—such as markets and recommendation systems—arising from shared states (e.g., prices, algorithmic recommendations), introducing the novel framework of “shared-state interference.” It provides the first formal characterization of this interference structure and proposes the key identification assumption of “conditional independence given the shared state.” Building upon this, the paper extends double machine learning (DML) theory to derive the semiparametric efficiency bound, orthogonal moment conditions, and identifiability criteria for this setting. It further develops estimators for the average direct effect (ADE) and the global average treatment effect (GATE) by flexibly modeling nuisance components using random forests, neural networks, or other modern ML methods. The resulting estimators are efficient, debiased, and asymptotically normal. The method substantially enhances the reliability and practicality of causal inference in interactive, shared-state systems.

Technology Category

• Machine Learning: Causal Learning
• Intelligent Robots: State Estimation
• Multiagent Systems: Mechanism Design

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

Researchers and practitioners often wish to measure treatment effects in settings where units interact via markets and recommendation systems. In these settings, units are affected by certain shared states, like prices, algorithmic recommendations or social signals. We formalize this structure, calling it shared-state interference, and argue that our formulation captures many relevant applied settings. Our key modeling assumption is that individuals' potential outcomes are independent conditional on the shared state. We then prove an extension of a double machine learning (DML) theorem providing conditions for achieving efficient inference under shared-state interference. We also instantiate our general theorem in several models of interest where it is possible to efficiently estimate the average direct effect (ADE) or global average treatment effect (GATE).

Problem

• Estimating treatment effects under shared-state interference in interactive systems
• Extending double machine learning for causal inference with shared states
• Providing efficient inference conditions for ADE and GATE estimation

Innovation

• Double Machine Learning for causal inference
• Shared-state interference modeling
• Efficient estimation of treatment effects