Reinforcement Learning for Robotic Control

Course Description

Reinforcement learning, a subset of artificial intelligence, is a computational approach that models decision-making by exploring the cause-and-effect relationships between actions and rewards. It is based on the idea that a system can learn through trial and error. Closely related to both optimal and adaptive control, reinforcement learning has significant applications in control systems. Reinforcement learning provides another framework for solving control problems where an agent interacts with its environment and refines its policies over time. In this course, we will introduce the fundamentals of reinforcement learning, Markov decision processes, dynamic programming, model-free prediction and control, value function approximation, and policy gradient methods. Then we will utilize reinforcement learning for robotic control. This course explores the fundamental principles of reinforcement learning and its integration into robotic control applications.

Code

ECE 5117

Credits

3

Subject

ECE