Counterfactual prediction is a fundamental task in decision-making. G-computation is a method for estimating expected counterfactual outcomes under dynamic time-varying treatment strategies. Existing G-computation implementations have mostly employed classical regression models with limited capacity to capture complex temporal and nonlinear dependence structures. This paper introduces G-Net, a novel sequential deep learning framework for G-computation that can handle complex time series data while imposing minimal modeling assumptions and provide estimates of individual or population-level time varying treatment effects. We evaluate alternative G-Net implementations using realistically complex temporal simulated data obtained from CVSim, a mechanistic model of the cardiovascular system.
@article{arxiv.2003.10551,
title = {G-Net: A Deep Learning Approach to G-computation for Counterfactual Outcome Prediction Under Dynamic Treatment Regimes},
author = {Rui Li and Zach Shahn and Jun Li and Mingyu Lu and Prithwish Chakraborty and Daby Sow and Mohamed Ghalwash and Li-wei H. Lehman},
journal= {arXiv preprint arXiv:2003.10551},
year = {2020}
}