Full Lecture on Reinforcement Learning
Full Lecture on Reinforcement Learning#
{"description": "Full-length lecture on reinforcement learning in Duckietown.", "keywords": "reinforcement learning, machine learning, ML, RL, policies, duckietown, AI, embedded AI, robotics"}
This lecture contains presents the material on Markov Decision Processes, Policy Iteration, Q Learning, and Simulation and Sim-to-real, as well as an additional section on policy gradients, an approach to learning an RL policy that does not explicitly estimate a value function.
This is another version of the slides on RL with more of a focus on model-based RL.