3.0

Average Course Rating

Tremendous success of reinforcement learning (RL) in a variety of settings from AlphaGo to LLMs makes it a critical area to study. This course will study classical aspects of RL as well as its modern counterparts. Topics will include Markov Decision Processes, dynamic programming, model-based and model-free RL, temporal difference learning, Monte Carlo methods, multi-armed bandits, policy optimization and other methods. Required course background: machine learning, linear algebra and probability. Students may receive credit for at most one of 601.479/679.