# Deep Doubly Debiased Longitudinal Effect Estimation with ICE G-Computation

arXiv: https://arxiv.org/abs/2602.12379 · HTML: https://mlsanity.com/paper/2602.12379/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 75 references that are in the map.
- First posted on arXiv 12 Feb 2026.

## Builds on (most cited first, 27 of 27)

- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Doubly Robust Off-policy Value Evaluation for Reinforcement Learning](https://mlsanity.com/paper/1511.03722/) (2015)
- [Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning](https://mlsanity.com/paper/1604.00923/) (2016)
- [Decoupling Representation Learning from Reinforcement Learning](https://mlsanity.com/paper/2009.08319/) (2020)
- [Batch Policy Learning under Constraints](https://mlsanity.com/paper/1903.08738/) (2019)
- [Breaking the Curse of Horizon: Infinite-Horizon Off-Policy Estimation](https://mlsanity.com/paper/1810.12429/) (2018)
- [Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations](https://mlsanity.com/paper/2002.04083/) (2020)
- [MIMIC-Extract: A Data Extraction, Preprocessing, and Representation Pipeline for MIMIC-III](https://mlsanity.com/paper/1907.08322/) (2019)
- [Empirical Study of Off-Policy Policy Evaluation for Reinforcement Learning](https://mlsanity.com/paper/1911.06854/) (2019)
- [Minimax-Optimal Off-Policy Evaluation with Linear Function Approximation](https://mlsanity.com/paper/2002.09516/) (2020)
- [Double Reinforcement Learning for Efficient Off-Policy Evaluation in Markov Decision Processes](https://mlsanity.com/paper/1908.08526/) (2019)
- [Importance Sampling Policy Evaluation with an Estimated Behavior Policy](https://mlsanity.com/paper/1806.01347/) (2018)
- [Estimating average causal effects from patient trajectories](https://mlsanity.com/paper/2203.01228/) (2022)
- [Double/Debiased Machine Learning for Dynamic Treatment Effects via g-Estimation](https://mlsanity.com/paper/2002.07285/) (2020)
- [Importance Resampling for Off-policy Prediction](https://mlsanity.com/paper/1906.04328/) (2019)
- [Model-agnostic meta-learners for estimating heterogeneous treatment effects over time](https://mlsanity.com/paper/2407.05287/) (2024)
- [More Efficient Off-Policy Evaluation through Regularized Targeted Learning](https://mlsanity.com/paper/1912.06292/) (2019)
- [Non-parametric causal effects based on longitudinal modified treatment policies](https://mlsanity.com/paper/2006.01366/) (2020)
- [Evaluating (weighted) dynamic treatment effects by double machine learning](https://mlsanity.com/paper/2012.00370/) (2020)
- [Sequential Double Robustness in Right-Censored Longitudinal Models](https://mlsanity.com/paper/1705.02459/) (2017)
- [Longitudinal Targeted Minimum Loss-based Estimation with Temporal-Difference Heterogeneous Transformer](https://mlsanity.com/paper/2404.04399/) (2024)
- [IGC-Net for conditional average potential outcome estimation over time](https://mlsanity.com/paper/2405.21012/) (2024)
- [Predictive auxiliary objectives in deep RL mimic learning in the brain](https://mlsanity.com/paper/2310.06089/) (2023)
- [Kernel Debiased Plug-in Estimation: Simultaneous, Automated Debiasing without Influence Functions for Many Target Parameters](https://mlsanity.com/paper/2306.08598/) (2023)
- [Deep Learning Methods for the Noniterative Conditional Expectation G-Formula for Causal Inference from Complex Observational Data](https://mlsanity.com/paper/2410.21531/) (2024)
- [Estimating the treatment effect over time under general interference through deep learner integrated TMLE](https://mlsanity.com/paper/2412.04799/) (2024)

## Abstract (arXiv)

Estimating longitudinal treatment effects is essential for sequential decision-making but is challenging due to treatment-confounder feedback. While Iterative Conditional Expectation (ICE) G-computation offers a principled approach, its recursive structure suffers from error propagation, corrupting the learned outcome regression models. We propose D3-Net, a framework that mitigates error propagation in ICE training and then applies a robust final correction. First, to interrupt error propagation during learning, we train the ICE sequence using Sequential Doubly Robust (SDR) pseudo-outcomes, which provide bias-corrected targets for each regression. Second, we employ a multi-task transformer with a covariate simulator head for auxiliary supervision, regularizing representation learning, and a target network to stabilize training dynamics. For the final estimate, we discard the SDR correction and instead use the uncorrected nuisance models to perform Longitudinal Targeted Minimum Loss-Based Estimation (LTMLE) on the original outcomes. This second-stage, targeted debiasing ensures robustness and optimal finite-sample properties. Comprehensive experiments demonstrate that our model, D3-Net, robustly reduces bias and variance across different horizons, counterfactuals, and time-varying confoundings, compared to existing state-of-the-art ICE-based estimators.