Hide Advanced Options
Courses - Fall 2026
MSAI
Master of Science in Artificial Intelligence
Open Seats as of
06/09/2026 at 01:30 PM
MSAI635
Reinforcement Learning
Credits: 3
Grad Meth: Reg, Aud
Prerequisite: MSAI601 and MSAI603.
Provides a broad introduction to reinforcement learning and sequential decision-making methods. Topics include: Foundations of reinforcement learning: Markov decision processes, Bellman equations, dynamic programming, policy evaluation and improvement, Monte Carlo methods, temporal-difference learning, Q-learning, and SARSA. Function approximation and deep reinforcement learning: value function approximation, deep Q-networks, policy gradient methods, actor-critic architectures, and model-based reinforcement learning. Additional topics may include multi-agent reinforcement learning, imitation learning, inverse reinforcement learning, exploration versus exploitation strategies, hierarchical reinforcement learning, and safe reinforcement learning. The course will also discuss recent applications of reinforcement learning, such as robotics, autonomous navigation, game playing, intelligent control systems, resource optimization, adaptive decision-making systems, and applications to training of large language models.