Minecraft Environments

Agent training and evaluation platforms built on the game Minecraft, used for open-world research.

Minecraft is an open-world sandbox game with long task chains and open-ended goals that often require crafting items through several steps, which is why it's commonly used as a virtual testbed for agents. MineRL, released in 2019 by a team including Carnegie Mellon University, provides Gym-style environments and over 60 million human demonstration state-action pairs, and has run competitions at NeurIPS, with its classic task being mining a diamond from scratch. MineDojo, released in 2022 by teams from NVIDIA, Caltech, Stanford, and others, contains thousands of open-ended tasks described in language along with a knowledge base built from 730,000 YouTube videos, wiki pages, and Reddit posts, using a video-language model called MineCLIP as a reward function; it won a NeurIPS 2022 Outstanding Paper Award.

ExampleVoyager has GPT-4 automatically write code and build up a skill library inside the MineDojo environment, continuously unlocking new tools and items.

Also called

MineDojo, MineRL

Related

VPT · Voyager · DreamerV3 · Open-world · Long-horizon Task · Reward Function

Sources

MineDojo 官网 (Chinese)
MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge (arXiv 2206.08853)
MineRL: A Large-Scale Dataset of Minecraft Demonstrations (arXiv 1907.13440)

As of

2022-11