Counterfactual Prediction with Deep Instrumental Variables Networks

By Jason S. Hartford, Greg Lewis, Kevin Leyton-Brown, Matt Taddy. 2016

48 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Causal Inference & Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Causal Inference & Discovery

Citations in the corpus

Machine Learning Methods That Economists Should Know About (2019, methodology): “A key method is the k-means algorithm (Hartigan and Wong [1979], Alpaydin [2009]).”
Monte Carlo Gradient Estimation in Machine Learning (2019, background): “…Kingma and Welling (2014b); Goodfellow et al. (2014); Bengio et al. (2014)), in bandits and online learning (Hu et al., 2016), in econometrics and simulation-based estimation (Gourieroux et al., 1996), and in instrumental-variables estimation and counter-factual reasoning (Hartford et al., 2016).”
Estimating individual treatment effect: generalization bounds and algorithms (2016, methodology): “Neural nets have also been used for this purpose, exemplified in early work by Beck et al. (2000), and more recently by Hartford et al. (2016)'s work on deep instrumental variables .”
How to make causal inferences using texts (2018, unknown): “There is also a burgeoning literature on causal inference using machine learning (Van der Laan and Rose, 2011; Athey, 2015; Athey and Imbens, 2016; Bloniarz et al., 2016; Hartford et al., 2016; Chernozhukov et al., 2017; Wager and Athey, 2017; Ratkovic and Tingley, 2017).”
Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes (2017, unknown): “[25] uses deep networks to infer counterfactuals, but requires some of the data to be drawn from a randomized trial.”