Hybrid RL + Imitation Learning for Robotics with Ray at RAI Institute

Learn how to implement hybrid reinforcement learning and imitation learning approaches for robotics applications using Ray framework in this 32-minute conference talk from Ray Summit 2025. Discover how RAI Institute leverages Ray's distributed computing capabilities to combine reinforcement learning with imitation learning techniques for training robotic systems. Explore the practical implementation of hybrid learning methodologies that can accelerate robot training by utilizing both reward-based learning and expert demonstrations. Understand how Ray's scalable infrastructure enables efficient training of complex robotic behaviors across distributed computing environments. Gain insights into the specific challenges of robotics applications and how hybrid learning approaches can overcome limitations of using either reinforcement learning or imitation learning in isolation. See real-world examples of how this methodology is being applied at RAI Institute to develop more capable and efficient robotic systems.