Games & virtual worlds · PROJECT NOTES
XLand-MiniGrid
A JAX grid-world suite for meta-reinforcement learning where composable rules and goals change the dynamics an agent must discover.
What can you do with it?
Train and evaluate adaptation to changing rulesets, using explicit train/test task distributions and optional visual observations.
What should you know before using it?
This is a from-scratch meta-learning suite inspired by XLand/MiniGrid, not a release of DeepMind's original XLand. First benchmark loading downloads cached ruleset data. PyPI 0.9.3 requires Python 3.9+ and appropriate JAX setup. Included MiniGrid ports overlap existing tasks and have limited maintenance. Code is Apache 2.0.
What is verified here?
This entry was reviewed against original project material. We have not independently run the software or reproduced its results. AI-development claims are author-reported; check upstream for current versions, setup requirements, and licenses.
- Resource type
- Environment + benchmark
- Provenance
- Development provenance not established
- Reviewed
- Platforms
- Consult upstream documentation