Neural MMO 2.0: A Massively Multi-task Addition to Massively Multi-agent Learning
This provides a new platform for reinforcement learning research, enabling scalable multi-agent experiments with customizable tasks, though it is incremental as an update to an existing environment.
The authors introduced Neural MMO 2.0, a massively multi-agent reinforcement learning environment with a flexible task system, challenging agents to generalize to unseen tasks, maps, and opponents, and reported a three-fold performance improvement over its predecessor.
Neural MMO 2.0 is a massively multi-agent environment for reinforcement learning research. The key feature of this new version is a flexible task system that allows users to define a broad range of objectives and reward signals. We challenge researchers to train agents capable of generalizing to tasks, maps, and opponents never seen during training. Neural MMO features procedurally generated maps with 128 agents in the standard setting and support for up to. Version 2.0 is a complete rewrite of its predecessor with three-fold improved performance and compatibility with CleanRL. We release the platform as free and open-source software with comprehensive documentation available at neuralmmo.github.io and an active community Discord. To spark initial research on this new platform, we are concurrently running a competition at NeurIPS 2023.