Counterfactual Prediction with Deep Instrumental Variables Networks

Abstract

We are in the middle of a remarkable rise in the use and capability of artificial intelligence. Much of this growth has been fueled by the success of deep learning architectures: models that map from observables to outputs via multiple layers of latent representations. These deep learning algorithms are effective tools for unstructured prediction, and they can be combined in AI systems to solve complex automated reasoning problems. This paper provides a recipe for combining ML algorithms to solve for causal effects in the presence of instrumental variables -- sources of treatment randomization that are conditionally independent from the response. We show that a flexible IV specification resolves into two prediction tasks that can be solved with deep neural nets: a first-stage network for treatment prediction and a second-stage network whose loss function involves integration over the conditional treatment distribution. This Deep IV framework imposes some specific structure on the stochastic gradient descent routine used for training, but it is general enough that we can take advantage of off-the-shelf ML capabilities and avoid extensive algorithm customization. We outline how to obtain out-of-sample causal validation in order to avoid over-fit. We also introduce schemes for both Bayesian and frequentist inference: the former via a novel adaptation of dropout training, and the latter via a data splitting routine.

References in corpus (6)

• Deep Learning in Neural Networks: An Overview
• Recursive Partitioning for Heterogeneous Causal Effects
• Nonparametric methods for inference in the presence of instrumental variables
• Optimal Inference After Model Selection
• On the Testability of Causal Models with Latent and Instrumental Variables
• Double/Debiased Machine Learning for Treatment and Causal Parameters

Cited by in corpus (11)

• Estimating individual treatment effect: generalization bounds and algorithms
• Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes
• Monte Carlo Gradient Estimation in Machine Learning
• Machine Learning Methods Economists Should Know About
• A Deep Causal Inference Approach to Measuring the Effects of Forming Group Loans in Online Non-profit Microfinance Platform
• Perturbing Inputs to Prevent Model Stealing
• Counterfactual-based Incrementality Measurement in a Digital Ad-Buying Platform
• Change Surfaces for Expressive Multidimensional Changepoints and Counterfactual Prediction
• Learning causal effects from many randomized experiments using regularized instrumental variables
• A framework for massive scale personalized promotion
• Estimation of Discrete Choice Models: A Machine Learning Approach