Double Machine Learning for Causal Inference under Shared-State Interference

Chris Hays, Manish Raghavan

TL;DR: We formalize the problem of shared-state interference in causal inference, prove a debiased machine learning theorem for inference in this setting, and instantiate the theorem in several settings of interest.

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).

Lay Summary: Researchers and practitioners often wish to measure causal effects in recommendation systems or markets, where units interact via centralized information, prices or other *shared states*. We develop theory allowing for estimation of causal effects in these settings without imposing strict assumptions on the data generating process while preserving the asymptotic efficiency of our method. This allows for valid and efficient inference in important, socially salient settings that was not possible before.

Application-Driven Machine Learning: This submission is on Application-Driven Machine Learning.

Link To Code: https://github.com/johnchrishays/dml4ssi

Primary Area: General Machine Learning->Causality

Keywords: double machine learning; interference; causal inference

Submission Number: 4853