Reinforcement Learning | Offered at Georgia Tech as CS 8803

Course Summary

You should take this course if you have an interest in machine learning and the desire to engage with it from a theoretical perspective. Through a combination of classic papers and more recent work, you will explore automated decision-making from a computer-science perspective. You will examine efficient algorithms, where they exist, for single-agent and multi-agent planning as well as approaches to learning near-optimal decisions from experience. At the end of the course, you will replicate a result from a published paper in reinforcement learning.

Why Take This Course?

This course will prepare you to participate in the reinforcement learning research community. You will also have the opportunity to learn from two of the foremost experts in this field of research, Profs. Charles Isbell and Michael Littman.

Prerequisites and Requirements

Before taking this course, you should have taken a graduate-level machine-learning course and should have had some exposure to reinforcement learning from a previous course or seminar in computer science (students who have completed CS 7641 will be well prepared for this course).

Additionally, you will be programming extensively in Java during this course. If you are not familiar with Java, we recommend you review Udacity's Intro to Java Programming course materials to get up to speed beforehand.

See the Technology Requirements for using Udacity.

What Will I Learn?

Projects

P4: TRAIN A SMARTCAB TO DRIVE

A smartcab is a self-driving car from the not-so-distant future that ferries people from one arbitrary location to another. In this project, you will use reinforcement learning to train a smartcab how to drive.

• Reinforcement Learning Basics
• Introduction to BURLAP
• TD Lambda
• Convergence of Value and Policy Iteration
• Reward Shaping
• Exploration
• Generalization
• Partially Observable MDPs
• Options
• Topics in Game Theory
• Further Topics in RL Models