D4RL: diagnostic datasets and tasks for offline reinforcement learning

What changed

Designed datasets spanning narrow support, non-RL controllers, heterogeneous behavior quality and multitask trajectories, with diagnostic navigation and robotic-control tasks for evaluating offline learners.

Scientific scope

Designed datasets spanning narrow support, non-RL controllers, heterogeneous behavior quality and multitask trajectories, with diagnostic navigation and robotic-control tasks for evaluating offline learners. Original v1 is titled Datasets for Data-Driven Reinforcement Learning and describes v0 datasets; later D4RL revisions or Minari are not silently substituted. Does not invent the underlying simulators or offline RL; benchmark success is not demonstrated real-world safety or broad deployment.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Original v1 is titled Datasets for Data-Driven Reinforcement Learning and describes v0 datasets; later D4RL revisions or Minari are not silently substituted. Does not invent the underlying simulators or offline RL; benchmark success is not demonstrated real-world safety or broad deployment.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Datasets for Data-Driven Reinforcement Learning

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction