About this game
ML Agents Walker is a browser-hosted Unity reinforcement-learning demo where movement itself is the game. It fits AI Game Lib as a simulation entry because the player watches embodied agents solve balance and locomotion problems in real time. Treat it like an AI behavior exhibit: observe stability, recovery, and the small failures that make trained motion systems readable.
ML Agents Walker sits in the AI Simulation Games section. The main appeal is watching a system respond over time, then making small changes to see how the outcome shifts. The tags #ml-agents, #simulation, #locomotion describe the fastest way to understand what kind of session to expect.
ML Agents Walker can run directly inside AI Game Lib, which makes it suitable for quick testing without leaving the directory.
