# The Edge-of-Reach Problem in Offline Model-Based Reinforcement Learning

arXiv: https://arxiv.org/abs/2402.12527 · HTML: https://mlsanity.com/paper/2402.12527/

## Key facts

- Cited by 8 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 29 of its 39 references that are in the map.
- First posted on arXiv 19 Feb 2024.

## Placed in

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

## Builds on (most cited first, 29 of 29)

- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [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)
- [Deep Reinforcement Learning with Double Q-learning](https://mlsanity.com/paper/1509.06461/) (2015)
- [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)
- [A Minimalist Approach to Offline Reinforcement Learning](https://mlsanity.com/paper/2106.06860/) (2021)
- [Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction](https://mlsanity.com/paper/1906.00949/) (2019)
- [Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models](https://mlsanity.com/paper/1805.12114/) (2018)
- [When to Trust Your Model: Model-Based Policy Optimization](https://mlsanity.com/paper/1906.08253/) (2019)
- [MOPO: Model-based Offline Policy Optimization](https://mlsanity.com/paper/2005.13239/) (2020)
- [MOReL : Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2005.05951/) (2020)
- [COMBO: Conservative Offline Model-Based Policy Optimization](https://mlsanity.com/paper/2102.08363/) (2021)
- [Uncertainty-Based Offline Reinforcement Learning with Diversified Q-Ensemble](https://mlsanity.com/paper/2110.01548/) (2021)
- [Randomized Ensembled Double Q-Learning: Learning Fast Without a Model](https://mlsanity.com/paper/2101.05982/) (2021)
- [Offline Reinforcement Learning with Fisher Divergence Critic Regularization](https://mlsanity.com/paper/2103.08050/) (2021)
- [Actionable Models: Unsupervised Offline Reinforcement Learning of Robotic Skills](https://mlsanity.com/paper/2104.07749/) (2021)
- [CORL: Research-oriented Deep Offline Reinforcement Learning Library](https://mlsanity.com/paper/2210.07105/) (2022)
- [RAMBO-RL: Robust Adversarial Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2204.12581/) (2022)
- [Offline Reinforcement Learning from Images with Latent Space Models](https://mlsanity.com/paper/2012.11547/) (2020)
- [Pre-Training for Robots: Offline RL Enables Learning New Tasks from a Handful of Trials](https://mlsanity.com/paper/2210.05178/) (2022)
- [A Workflow for Offline Model-Free Robotic Reinforcement Learning](https://mlsanity.com/paper/2109.10813/) (2021)
- [Model Selection for Offline Reinforcement Learning: Practical Considerations for Healthcare Settings](https://mlsanity.com/paper/2107.11003/) (2021)
- [Challenges and Opportunities in Offline Reinforcement Learning from Visual Observations](https://mlsanity.com/paper/2206.04779/) (2022)
- [Revisiting Design Choices in Offline Model-Based Reinforcement Learning](https://mlsanity.com/paper/2110.04135/) (2021)
- [Augmented World Models Facilitate Zero-Shot Dynamics Generalization From a Single Offline Environment](https://mlsanity.com/paper/2104.05632/) (2021)
- [Mitigating Value Hallucination in Dyna Planning via Multistep Predecessor Models](https://mlsanity.com/paper/2006.04363/) (2020)

## Built on by (most cited first, 8 of 8)

- [Policy-Guided Diffusion](https://mlsanity.com/paper/2404.06356/) (2024)
- [Scalable Offline Model-Based RL with Action Chunks](https://mlsanity.com/paper/2512.08108/) (2025)
- [Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2411.04562/) (2024)
- [Predicting Closed-Loop Performance of Latent World Models: Offline Checkpoint Selection for MPC and Model-Based RL Under Non-Markovian Rewards in LunarLander](https://mlsanity.com/paper/2607.01736/) (2026)
- [Offline vs. Online Learning in Model-based RL: Lessons for Data Collection Strategies](https://mlsanity.com/paper/2509.05735/) (2025)
- [Long-Horizon Model-Based Offline Reinforcement Learning Without Explicit Conservatism](https://mlsanity.com/paper/2512.04341/) (2025)
- [Beyond Static Datasets: Robust Offline Policy Optimization via Vetted Synthetic Transitions](https://mlsanity.com/paper/2601.18107/) (2026)
- [From Static Policies to Adaptive Priors in Offline Reinforcement Learning](https://mlsanity.com/paper/2609.35880/) (2026)

## Abstract (arXiv)

Offline reinforcement learning aims to train agents from pre-collected datasets. However, this comes with the added challenge of estimating the value of behaviors not covered in the dataset. Model-based methods offer a potential solution by training an approximate dynamics model, which then allows collection of additional synthetic data via rollouts in this model. The prevailing theory treats this approach as online RL in an approximate dynamics model, and any remaining performance gap is therefore understood as being due to dynamics model errors. In this paper, we analyze this assumption and investigate how popular algorithms perform as the learned dynamics model is improved. In contrast to both intuition and theory, if the learned dynamics model is replaced by the true error-free dynamics, existing model-based methods completely fail. This reveals a key oversight: The theoretical foundations assume sampling of full horizon rollouts in the learned dynamics model; however, in practice, the number of model-rollout steps is aggressively reduced to prevent accumulating errors. We show that this truncation of rollouts results in a set of edge-of-reach states at which we are effectively ``bootstrapping from the void.'' This triggers pathological value overestimation and complete performance collapse. We term this the edge-of-reach problem. Based on this new insight, we fill important gaps in existing theory, and reveal how prior model-based methods are primarily addressing the edge-of-reach problem, rather than model-inaccuracy as claimed. Finally, we propose Reach-Aware Value Learning (RAVL), a simple and robust method that directly addresses the edge-of-reach problem and hence - unlike existing methods - does not fail as the dynamics model is improved. Code open-sourced at: github.com/anyasims/edge-of-reach.