A Primer on Deep Learning for Causal Inference
- 16 August 2024
- journal article
- research article
- Published by SAGE Publications in Sociological Methods & Research
- Vol. 54 (2) , 397-447
Abstract
This primer systematizes the emerging literature on causal inference using deep neural networks under the potential outcomes framework. It provides an intuitive introduction to building and optimizing custom deep learning models and shows how to adapt them to estimate/predict heterogeneous treatment effects. It also discusses ongoing work to extend causal inference to settings where confounding is nonlinear, time-varying, or encoded in text, networks, and images. To maximize accessibility, we also introduce prerequisite concepts from causal inference and deep learning. The primer differs from other treatments of deep learning and causal inference in its sharp focus on observational causal estimation, its extended exposition of key algorithms, and its detailed tutorials for implementing, training, and selecting among deep estimators in TensorFlow 2 and PyTorch.Keywords
All Related Versions
Funding Information
- Agencia Nacional de Investigación y Desarrollo (2018-72190240)
- National Science Foundation (2124685, Graduate Research Fellowship and Doctoral Disserta)
This publication has 50 references indexed in Scilit:
- The Principles of Deep Learning Theory2022
- Semiparametric Theory and Empirical Processes in Causal Inference2016
- Deep Learning of Representations: Looking Forward2013
- Targeted Learning2011
- The Elements of Statistical Learning2009
- Breastfeeding and Child Cognitive DevelopmentArchives of General Psychiatry, 2008
- Qualitative probabilities for default reasoning, belief revision, and causal modelingArtificial Intelligence, 1996
- Approximation by superpositions of a sigmoidal functionMathematics of Control, Signals, and Systems, 1992
- A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periodsJournal of Chronic Diseases, 1987
- A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effectMathematical Modelling, 1986