# Deconfounded Warm-Start Thompson Sampling with Applications to Precision Medicine

arXiv: https://arxiv.org/abs/2505.17283 · HTML: https://mlsanity.com/paper/2505.17283/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 47 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 23 of 23)

- [Estimation and Inference of Heterogeneous Treatment Effects using Random Forests](https://mlsanity.com/paper/1510.04342/) (2015)
- [Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning](https://mlsanity.com/paper/1706.03461/) (2017)
- [Causal Effect Inference with Deep Latent-Variable Models](https://mlsanity.com/paper/1705.08821/) (2017)
- [Adapting Neural Networks for the Estimation of Treatment Effects](https://mlsanity.com/paper/1906.02120/) (2019)
- [Nonparametric Estimation of Heterogeneous Treatment Effects: From Theory to Learning Algorithms](https://mlsanity.com/paper/2101.10943/) (2021)
- [The Blessings of Multiple Causes](https://mlsanity.com/paper/1805.06826/) (2018)
- [Removing Hidden Confounding by Experimental Grounding](https://mlsanity.com/paper/1810.11646/) (2018)
- [Leveraging Offline Data in Online Reinforcement Learning](https://mlsanity.com/paper/2211.04974/) (2022)
- [Combining Observational and Randomized Data for Estimating Heterogeneous Treatment Effects](https://mlsanity.com/paper/2202.12891/) (2022)
- [Combining multiple observational data sources to estimate causal effects](https://mlsanity.com/paper/1801.00802/) (2018)
- [Combining Observational and Experimental Datasets Using Shrinkage Estimators](https://mlsanity.com/paper/2002.06708/) (2020)
- [Estimating Average Treatment Effects via Orthogonal Regularization](https://mlsanity.com/paper/2101.08490/) (2021)
- [Bandits with Partially Observable Confounded Data](https://mlsanity.com/paper/2006.06731/) (2020)
- [Adaptive Combination of Randomized and Observational Data](https://mlsanity.com/paper/2111.15012/) (2021)
- [Warm-starting Contextual Bandits: Robustly Combining Supervised and Bandit Feedback](https://mlsanity.com/paper/1901.00301/) (2019)
- [Deconfounding Temporal Autoencoder: Estimating Treatment Effects over Time Using Noisy Proxies](https://mlsanity.com/paper/2112.03013/) (2021)
- [Artificial Replay: A Meta-Algorithm for Harnessing Historical Data in Bandits](https://mlsanity.com/paper/2210.00025/) (2022)
- [Leveraging Demonstrations to Improve Online Learning: Quality Matters](https://mlsanity.com/paper/2302.03319/) (2023)
- [Estimating heterogeneous survival treatment effect in observational data using machine learning](https://mlsanity.com/paper/2008.07044/) (2020)
- [Doubly Debiased Lasso: High-Dimensional Inference under Hidden Confounding](https://mlsanity.com/paper/2004.03758/) (2020)
- [Linear Bandits with Partially Observable Features](https://mlsanity.com/paper/2502.06142/) (2025)
- [Efficient Online Learning with Offline Datasets for Infinite Horizon MDPs: A Bayesian Approach](https://mlsanity.com/paper/2310.11531/) (2023)
- [On LASSO for High Dimensional Predictive Regression](https://mlsanity.com/paper/2212.07052/) (2022)

## Abstract (arXiv)

Randomized clinical trials often require large patient cohorts before drawing definitive conclusions, yet abundant observational data from parallel studies remains underutilized due to confounding and hidden biases. To bridge this gap, we propose Deconfounded Warm-Start Thompson Sampling (DWTS), a practical approach that leverages a Doubly Debiased LASSO (DDL) procedure to identify a sparse set of reliable measured covariates and combines them with key hidden covariates to form a reduced context. By initializing Thompson Sampling (LinTS) priors with DDL-estimated means and variances on these measured features -- while keeping uninformative priors on hidden features -- DWTS effectively harnesses confounded observational data to kick-start adaptive clinical trials. Evaluated on both a purely synthetic environment and a virtual environment created using real cardiovascular risk dataset, DWTS consistently achieves lower cumulative regret than standard LinTS, showing how offline causal insights from observational data can improve trial efficiency and support more personalized treatment decisions.