# G-Net: A Deep Learning Approach to G-computation for Counterfactual Outcome Prediction Under Dynamic Treatment Regimes

arXiv: https://arxiv.org/abs/2003.10551 · HTML: https://mlsanity.com/paper/2003.10551/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Mar 2020.

## Built on by (most cited first, 13 of 13)

- [Causal Transformer for Estimating Counterfactual Outcomes](https://mlsanity.com/paper/2204.07258/) (2022)
- [Causal machine learning for predicting treatment outcomes](https://mlsanity.com/paper/2410.08770/) (2024)
- [Accounting For Informative Sampling When Learning to Forecast Treatment Outcomes Over Time](https://mlsanity.com/paper/2306.04255/) (2023)
- [A Survey of Deep Causal Models and Their Industrial Applications](https://mlsanity.com/paper/2209.08860/) (2022)
- [Deep Learning for Time Series Forecasting: A Survey](https://mlsanity.com/paper/2503.10198/) (2025)
- [Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimation](https://mlsanity.com/paper/2506.21154/) (2025)
- [Training and Evaluating Causal Forecasting Models for Time-Series](https://mlsanity.com/paper/2411.00126/) (2024)
- [Beyond the Known: Decision Making with Counterfactual Reasoning Decision Transformer](https://mlsanity.com/paper/2505.09114/) (2025)
- [Causal Diffusion Models for Counterfactual Outcome Distributions in Longitudinal Data](https://mlsanity.com/paper/2604.12992/) (2026)
- [Inferring Effects of Major Events through Discontinuity Forecasting of Population Anxiety](https://mlsanity.com/paper/2508.21722/) (2025)
- [Synergizing Deconfounding and Temporal Generalization For Time-series Counterfactual Outcome Estimation](https://mlsanity.com/paper/2511.16006/) (2025)
- [Observable Neural ODEs for Identifiable Causal Forecasting in Continuous Time](https://mlsanity.com/paper/2604.26070/) (2026)
- [From Static Risk to Dynamic Trajectories: Toward World-Model-Inspired Clinical Prediction](https://mlsanity.com/paper/2605.16927/) (2026)

## Abstract (arXiv)

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.