# Offline Policy Evaluation and Optimization under Confounding

arXiv: https://arxiv.org/abs/2211.16583 · HTML: https://mlsanity.com/paper/2211.16583/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Offline RL algorithms, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 24 of its 51 references that are in the map.
- First posted on arXiv 29 Nov 2022.

## Placed in

- Offline RL and imitation learning from data / [Offline RL algorithms](https://mlsanity.com/offlinerl/offline-rl-algorithms/)

## Builds on (most cited first, 24 of 24)

- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [Minimax-Optimal Off-Policy Evaluation with Linear Function Approximation](https://mlsanity.com/paper/2002.09516/) (2020)
- [Counterfactual Off-Policy Evaluation with Gumbel-Max Structural Causal Models](https://mlsanity.com/paper/1905.05824/) (2019)
- [The Limit Points of (Optimistic) Gradient Descent in Min-Max Optimization](https://mlsanity.com/paper/1807.03907/) (2018)
- [Confounding-Robust Policy Improvement](https://mlsanity.com/paper/1805.08593/) (2018)
- [RL for Latent MDPs: Regret Guarantees and a Lower Bound](https://mlsanity.com/paper/2102.04939/) (2021)
- [Off-Policy Evaluation in Partially Observable Environments](https://mlsanity.com/paper/1909.03739/) (2019)
- [Off-policy Policy Evaluation For Sequential Decisions Under Unobserved Confounding](https://mlsanity.com/paper/2003.05623/) (2020)
- [Asymptotically Efficient Off-Policy Evaluation for Tabular Reinforcement Learning](https://mlsanity.com/paper/2001.10742/) (2020)
- [Confounding-Robust Policy Evaluation in Infinite-Horizon Reinforcement Learning](https://mlsanity.com/paper/2002.04518/) (2020)
- [Off-policy Evaluation in Infinite-Horizon Reinforcement Learning with Latent Confounders](https://mlsanity.com/paper/2007.13893/) (2020)
- [Provably Efficient Causal Reinforcement Learning with Confounded Observational Data](https://mlsanity.com/paper/2006.12311/) (2020)
- [A Minimax Learning Approach to Off-Policy Evaluation in Confounded Partially Observable Markov Decision Processes](https://mlsanity.com/paper/2111.06784/) (2021)
- [Instrumental Variable Value Iteration for Causal Offline Reinforcement Learning](https://mlsanity.com/paper/2102.09907/) (2021)
- [Pessimism in the Face of Confounders: Provably Efficient Offline Reinforcement Learning in Partially Observable Markov Decision Processes](https://mlsanity.com/paper/2205.13589/) (2022)
- [Gradient Descent Converges to Minimizers](https://mlsanity.com/paper/1602.04915/) (2016)
- [Statistically Efficient Off-Policy Policy Gradients](https://mlsanity.com/paper/2002.04014/) (2020)
- [A Spectral Approach to Off-Policy Evaluation for POMDPs](https://mlsanity.com/paper/2109.10502/) (2021)
- [Future-Dependent Value-Based Off-Policy Evaluation in POMDPs](https://mlsanity.com/paper/2207.13081/) (2022)
- [Offline Reinforcement Learning with Instrumental Variables in Confounded Markov Decision Processes](https://mlsanity.com/paper/2209.08666/) (2022)
- [Off-Policy Fitted Q-Evaluation with Differentiable Function Approximators: Z-Estimation and Inference Theory](https://mlsanity.com/paper/2202.04970/) (2022)
- [Off-Policy Evaluation for Episodic Partially Observable Markov Decision Processes under Non-Parametric Models](https://mlsanity.com/paper/2209.10064/) (2022)
- [Model-Free and Model-Based Policy Evaluation when Causality is Uncertain](https://mlsanity.com/paper/2204.00956/) (2022)
- [Learning Mixtures of Markov Chains and MDPs](https://mlsanity.com/paper/2211.09403/) (2022)

## Built on by (most cited first, 12 of 12)

- [Confounding Robust Deep Reinforcement Learning: A Causal Approach](https://mlsanity.com/paper/2510.21110/) (2025)
- [Efficient and Sharp Off-Policy Learning under Unobserved Confounding](https://mlsanity.com/paper/2502.13022/) (2025)
- [A Natural Extension To Online Algorithms For Hybrid RL With Limited Coverage](https://mlsanity.com/paper/2403.09701/) (2024)
- [Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning](https://mlsanity.com/paper/2412.05783/) (2024)
- [Offline Recommender System Evaluation under Unobserved Confounding](https://mlsanity.com/paper/2309.04222/) (2023)
- [Why Online Reinforcement Learning is Causal](https://mlsanity.com/paper/2403.04221/) (2024)
- [Off-Policy Evaluation for Sequential Persuasion Process with Unobserved Confounding](https://mlsanity.com/paper/2504.01211/) (2025)
- [Model-Based Reinforcement Learning Under Confounding](https://mlsanity.com/paper/2512.07528/) (2025)
- [Unifying Causal Reinforcement Learning: Survey, Taxonomy, Algorithms and Applications](https://mlsanity.com/paper/2512.18135/) (2025)
- [Pessimistic Causal Reinforcement Learning with Mediators for Confounded Offline Data](https://mlsanity.com/paper/2403.11841/) (2024)
- [Non-Parametric Rehearsal Learning via Conditional Mean Embeddings](https://mlsanity.com/paper/2605.08999/) (2026)
- [Robust Counterfactual Policy Optimisation via Nondeterministic Causal Models](https://mlsanity.com/paper/2608.02893/) (2026)

## Abstract (arXiv)

Evaluating and optimizing policies in the presence of unobserved confounders is a problem of growing interest in offline reinforcement learning. Using conventional methods for offline RL in the presence of confounding can not only lead to poor decisions and poor policies, but also have disastrous effects in critical applications such as healthcare and education. We map out the landscape of offline policy evaluation for confounded MDPs, distinguishing assumptions on confounding based on whether they are memoryless and on their effect on the data-collection policies. We characterize settings where consistent value estimates are provably not achievable, and provide algorithms with guarantees to instead estimate lower bounds on the value. When consistent estimates are achievable, we provide algorithms for value estimation with sample complexity guarantees. We also present new algorithms for offline policy improvement and prove local convergence guarantees. Finally, we experimentally evaluate our algorithms on both a gridworld environment and a simulated healthcare setting of managing sepsis patients. In gridworld, our model-based method provides tighter lower bounds than existing methods, while in the sepsis simulator, our methods significantly outperform confounder-oblivious benchmarks.