openai/multi-agent-emergence-environments
Simulation environments for training and studying emergent behavior in multi-agent systems through interactive scenarios like hide-and-seek and box locking.
Not currently ranked — collecting fresh signals.
star history
Provides simulation environments designed for multi-agent reinforcement learning research, supporting complex agent interactions like hide-and-seek and object manipulation. The environment construction system uses modular components (EnvModules and gym.Wrappers) that enable researchers to build and customize agent training scenarios. Supports configurable game variants including random rooms, quadrant, and food scenarios.
Frequently asked
- What is openai/multi-agent-emergence-environments?
- Simulation environments for training and studying emergent behavior in multi-agent systems through interactive scenarios like hide-and-seek and box locking.
- Is multi-agent-emergence-environments open source?
- Yes — openai/multi-agent-emergence-environments is open source, released under the MIT license.
- What language is multi-agent-emergence-environments written in?
- openai/multi-agent-emergence-environments is primarily written in Python.
- How popular is multi-agent-emergence-environments?
- openai/multi-agent-emergence-environments has 1.8k stars on GitHub.
- Where can I find multi-agent-emergence-environments?
- openai/multi-agent-emergence-environments is on GitHub at https://github.com/openai/multi-agent-emergence-environments.