Glossary
RLDS
Reinforcement Learning Datasets — an episodic format for robot learning data.
Definition
RLDS (Reinforcement Learning Datasets) is a standard for storing episodic, step-based data for robot learning, built on TFDS. It represents each demonstration as a sequence of steps with observations and actions, which makes egocentric episodes portable across training pipelines.
01
Why it matters
A common episodic format means you can mix datasets and reuse dataloaders instead of writing bespoke parsers. RLDS is widely used across open robot-learning efforts.
02
How Firsthand handles it
Firsthand delivers in RLDS with readable converter source, so you can retarget the field mapping to your own loader rather than trusting a binary.
Related terms