COE550 Reinforcement Learning

[3–0, 3 cr.]

This course introduces the core principles and algorithms of Reinforcement Learning, focusing on how agents learn optimal behavior through interaction with an environment to maximize cumulative rewards. Students will build a foundation in Markov decision processes, dynamic programming, Monte Carlo methods, and temporal-difference learning. The course also covers topics such as exploration strategies, multi-armed bandits, policy gradient methods, Deep Reinforcement Learning, and Reinforcement Learning from Human Feedback.

Pre-requisites: (GNE331 or MTH305) and (COE211 or COE212) and fourth-year standing.