# Exploiting Action Impact Regularity and Exogenous State Variables for Offline Reinforcement Learning

arXiv: https://arxiv.org/abs/2111.08066 · HTML: https://mlsanity.com/paper/2111.08066/

## Key facts

- Cited by 2 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 23 of its 54 references that are in the map.
- First posted on arXiv 15 Nov 2021.

## Placed in

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

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

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [Offline Reinforcement Learning with Implicit Q-Learning](https://mlsanity.com/paper/2110.06169/) (2021)
- [Off-Policy Deep Reinforcement Learning without Exploration](https://mlsanity.com/paper/1812.02900/) (2018)
- [Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction](https://mlsanity.com/paper/1906.00949/) (2019)
- [Behavior Regularized Offline Reinforcement Learning](https://mlsanity.com/paper/1911.11361/) (2019)
- [MOReL : Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2005.05951/) (2020)
- [Information-Theoretic Considerations in Batch Reinforcement Learning](https://mlsanity.com/paper/1905.00360/) (2019)
- [Safe Policy Improvement with Baseline Bootstrapping](https://mlsanity.com/paper/1712.06924/) (2017)
- [What are the Statistical Limits of Offline RL with Linear Function Approximation?](https://mlsanity.com/paper/2010.11895/) (2020)
- [Model-Based Reinforcement Learning with a Generative Model is Minimax Optimal](https://mlsanity.com/paper/1906.03804/) (2019)
- [Learning with Good Feature Representations in Bandits and in RL with a Generative Model](https://mlsanity.com/paper/1911.07676/) (2019)
- [Provably Good Batch Reinforcement Learning Without Great Exploration](https://mlsanity.com/paper/2007.08202/) (2020)
- [Exponential Lower Bounds for Batch Reinforcement Learning: Batch RL can be Exponentially Harder than Online RL](https://mlsanity.com/paper/2012.08005/) (2020)
- [Risk Bounds and Rademacher Complexity in Batch Reinforcement Learning](https://mlsanity.com/paper/2103.13883/) (2021)
- [Towards Hyperparameter-free Policy Selection for Offline Reinforcement Learning](https://mlsanity.com/paper/2110.14000/) (2021)
- [Provable RL with Exogenous Distractors via Multistep Inverse Dynamics](https://mlsanity.com/paper/2110.08847/) (2021)
- [Discovering and Removing Exogenous State Variables and Rewards for Reinforcement Learning](https://mlsanity.com/paper/1806.01584/) (2018)
- [Efficient Planning in Large MDPs with Weak Linear Function Approximation](https://mlsanity.com/paper/2007.06184/) (2020)
- [Learning Compact Models for Planning with Exogenous Processes](https://mlsanity.com/paper/1909.13870/) (2019)
- [Factoring Exogenous State for Model-Free Monte Carlo](https://mlsanity.com/paper/1703.09390/) (2017)

## Built on by (most cited first, 2 of 2)

- [Identifiability and generalizability from multiple experts in Inverse Reinforcement Learning](https://mlsanity.com/paper/2209.10974/) (2022)
- [Value Bonuses using Ensemble Errors for Exploration in Reinforcement Learning](https://mlsanity.com/paper/2602.12375/) (2026)

## Abstract (arXiv)

Offline reinforcement learning -- learning a policy from a batch of data -- is known to be hard for general MDPs. These results motivate the need to look at specific classes of MDPs where offline reinforcement learning might be feasible. In this work, we explore a restricted class of MDPs to obtain guarantees for offline reinforcement learning. The key property, which we call Action Impact Regularity (AIR), is that actions primarily impact a part of the state (an endogenous component) and have limited impact on the remaining part of the state (an exogenous component). AIR is a strong assumption, but it nonetheless holds in a number of real-world domains including financial markets. We discuss algorithms that exploit the AIR property, and provide a theoretical analysis for an algorithm based on Fitted-Q Iteration. Finally, we demonstrate that the algorithm outperforms existing offline reinforcement learning algorithms across different data collection policies in simulated and real world environments where the regularity holds.