Deep Dive: The Mujoko Fly—Biology, Simulation, and Reinforcement Learning

A concise, accessible tour of the Mujoko-based fruit fly model (Flybody) from Google DeepMind and Janelia Research Campus. We explore how anatomically detailed physics drives reinforcement learning for flight and walking, vision-guided tasks, and how these building blocks could inform robotics and embodied AI.

Note: This podcast was AI-generated, and sometimes AI can make mistakes. Please double-check any critical information.

Sponsored by Embersilk LLC