Offline Reinforcement Learning from Images with Latent Space Models

by   Rafael Rafailov, et al.

Offline reinforcement learning (RL) refers to the problem of learning policies from a static dataset of environment interactions. Offline RL enables extensive use and re-use of historical datasets, while also alleviating safety concerns associated with online exploration, thereby expanding the real-world applicability of RL. Most prior work in offline RL has focused on tasks with compact state representations. However, the ability to learn directly from rich observation spaces like images is critical for real-world applications such as robotics. In this work, we build on recent advances in model-based algorithms for offline RL, and extend them to high-dimensional visual observation spaces. Model-based offline RL algorithms have achieved state of the art results in state based tasks and have strong theoretical guarantees. However, they rely crucially on the ability to quantify uncertainty in the model predictions, which is particularly challenging with image observations. To overcome this challenge, we propose to learn a latent-state dynamics model, and represent the uncertainty in the latent space. Our approach is both tractable in practice and corresponds to maximizing a lower bound of the ELBO in the unknown POMDP. In experiments on a range of challenging image-based locomotion and manipulation tasks, we find that our algorithm significantly outperforms previous offline model-free RL methods as well as state-of-the-art online visual model-based RL methods. Moreover, we also find that our approach excels on an image-based drawer closing task on a real robot using a pre-existing dataset. All results including videos can be found online at .


page 8

page 20


MOReL : Model-Based Offline Reinforcement Learning

In offline reinforcement learning (RL), the goal is to learn a successfu...

Transfer RL across Observation Feature Spaces via Model-Based Regularization

In many reinforcement learning (RL) applications, the observation space ...

Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL

Offline Reinforcement Learning (RL) aims to extract near-optimal policie...

INFOrmation Prioritization through EmPOWERment in Visual Model-Based RL

Model-based reinforcement learning (RL) algorithms designed for handling...

Beyond Tabula-Rasa: a Modular Reinforcement Learning Approach for Physically Embedded 3D Sokoban

Intelligent robots need to achieve abstract objectives using concrete, s...

Extending Deep Model Predictive Control with Safety Augmented Value Estimation from Demonstrations

Reinforcement learning (RL) for robotics is challenging due to the diffi...

Recovery RL: Safe Reinforcement Learning with Learned Recovery Zones

Safety remains a central obstacle preventing widespread use of RL in the...

Code Repositories


Official Codebase for Offline Reinforcement Learning from Images with Latent Space Models

view repo