# Learning Optimal Distributionally Robust Individualized Treatment Rules

arXiv: https://arxiv.org/abs/2006.15121 · HTML: https://mlsanity.com/paper/2006.15121/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jun 2020.

## Built on by (most cited first, 28 of 28)

- [A Review of Off-Policy Evaluation in Reinforcement Learning](https://mlsanity.com/paper/2212.06355/) (2022)
- [Doubly Robust Distributionally Robust Off-Policy Evaluation and Learning](https://mlsanity.com/paper/2202.09667/) (2022)
- [Generalizing Off-Policy Learning under Sample Selection Bias](https://mlsanity.com/paper/2112.01387/) (2021)
- [Safe Policy Learning through Extrapolation: Application to Pre-trial Risk Assessment](https://mlsanity.com/paper/2109.11679/) (2021)
- [Median Optimal Treatment Regimes](https://mlsanity.com/paper/2103.01802/) (2021)
- [Robust Offline Reinforcement learning with Heavy-Tailed Rewards](https://mlsanity.com/paper/2310.18715/) (2023)
- [Minimax Regret Learning for Data with Heterogeneous Subgroups](https://mlsanity.com/paper/2405.01709/) (2024)
- [Efficient and robust transfer learning of optimal individualized treatment regimes with right-censored survival data](https://mlsanity.com/paper/2301.05491/) (2023)
- [Minimax Regret Estimation for Generalizing Heterogeneous Treatment Effects with Multisite Data](https://mlsanity.com/paper/2412.11136/) (2024)
- [A Multi-Agent Reinforcement Learning Framework for Off-Policy Evaluation in Two-sided Markets](https://mlsanity.com/paper/2202.10574/) (2022)
- [Decision-Focused Evaluation of Worst-Case Distribution Shift](https://mlsanity.com/paper/2407.03557/) (2024)
- [Learning Under Adversarial and Interventional Shifts](https://mlsanity.com/paper/2103.15933/) (2021)
- [Robust Sample Weighting to Facilitate Individualized Treatment Rule Learning for a Target Population](https://mlsanity.com/paper/2105.00581/) (2021)
- [STEEL: Singularity-aware Reinforcement Learning](https://mlsanity.com/paper/2301.13152/) (2023)
- [Practical Performative Policy Learning with Strategic Agents](https://mlsanity.com/paper/2412.01344/) (2024)
- [Targeted Optimal Treatment Regime Learning Using Summary Statistics](https://mlsanity.com/paper/2201.06229/) (2022)
- [Stage-Aware Learning for Dynamic Treatments](https://mlsanity.com/paper/2310.19300/) (2023)
- [Doubly Robust Fusion of Many Treatments for Policy Learning](https://mlsanity.com/paper/2505.08092/) (2025)
- [Calibrated Optimal Decision Making with Multiple Data Sources and Limited Outcome](https://mlsanity.com/paper/2104.10554/) (2021)
- [Distributionally Robust Learning in Survival Analysis](https://mlsanity.com/paper/2506.01348/) (2025)
- [Online Bayesian Risk-Averse Reinforcement Learning](https://mlsanity.com/paper/2509.14077/) (2025)
- [Learning Polyhedral Conformal Sets for Robust Optimization](https://mlsanity.com/paper/2605.08506/) (2026)
- [Rejoinder: Learning Optimal Distributionally Robust Individualized Treatment Rules](https://mlsanity.com/paper/2110.08936/) (2021)
- [Towards Robust Off-Policy Evaluation via Human Inputs](https://mlsanity.com/paper/2209.08682/) (2022)
- [Deep Spectral Q-learning with Application to Mobile Health](https://mlsanity.com/paper/2301.00927/) (2023)
- [Learning Optimal Distributionally Robust Individualized Treatment Rules Integrating Multi-Source Data](https://mlsanity.com/paper/2603.05568/) (2026)
- [Counterfactually Safe Reinforcement Learning](https://mlsanity.com/paper/2605.25114/) (2026)
- [Wasserstein Policy Learning for Distributional Outcomes](https://mlsanity.com/paper/2606.19117/) (2026)

## Abstract (arXiv)

Recent development in the data-driven decision science has seen great advances in individualized decision making. Given data with individual covariates, treatment assignments and outcomes, policy makers best individualized treatment rule (ITR) that maximizes the expected outcome, known as the value function. Many existing methods assume that the training and testing distributions are the same. However, the estimated optimal ITR may have poor generalizability when the training and testing distributions are not identical. In this paper, we consider the problem of finding an optimal ITR from a restricted ITR class where there is some unknown covariate changes between the training and testing distributions. We propose a novel distributionally robust ITR (DR-ITR) framework that maximizes the worst-case value function across the values under a set of underlying distributions that are "close" to the training distribution. The resulting DR-ITR can guarantee the performance among all such distributions reasonably well. We further propose a calibrating procedure that tunes the DR-ITR adaptively to a small amount of calibration data from a target population. In this way, the calibrated DR-ITR can be shown to enjoy better generalizability than the standard ITR based on our numerical studies.