ml-agents and ml-agents-dodgeball-env
ML-Agents is the core framework for training agents in Unity environments, while ml-agents-dodgeball-env is a specific example environment that demonstrates and depends on the ML-Agents toolkit, making them complements designed to be used together.
About ml-agents
Unity-Technologies/ml-agents
The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.
This toolkit helps game developers and researchers create intelligent characters and systems within Unity games and simulations. You provide a Unity game environment, and the toolkit outputs trained AI agents that can control Non-Player Characters (NPCs), automate game testing, or evaluate design choices. Game developers and AI researchers are the primary users.
About ml-agents-dodgeball-env
Unity-Technologies/ml-agents-dodgeball-env
Showcase environment for ML-Agents
This environment helps game developers or AI researchers create and test complex AI behaviors for cooperative and adversarial team-based games. You provide game environments and player actions, and it outputs trained AI agents capable of playing dodgeball-style games. Game AI designers and researchers focused on multi-agent learning would use this.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work