Recent news coverage has highlighted how reinforcement learning algorithms are now beating professionals in games like GO, Dota 2, and Starcraft 2. Reinforcement learning is a type of machine learning that enables the use of artificial intelligence in complex applications from video games to robotics, self-driving cars, and more.

If you are interested in using reinforcement learning technology for your project, but you’ve never used it before, where do you begin?

This ebook will help you get started with reinforcement learning in MATLAB® and Simulink® by explaining the terminology and providing access to examples, tutorials, and trial software.

Download this ebook to learn about:

Section 1: The Basics and Setting Up the Environment

Learn the basics of reinforcement learning and how it compares with traditional control design. See the difference between supervised, unsupervised, and reinforcement learning, and see how to set up a learning environment in MATLAB and Simulink. 

Section 2: Rewards and Policy Structures

Learn about exploration and exploitation in reinforcement learning and how to shape reward functions. Explore different options for representing policies including neural networks and how they can be used as function approximators.

Section 3: Understanding Training and Deployment

Learn about the different types of training algorithms, including policy-based, value-based and actor-critic methods. Find out more about the pros and cons of each training method as well as the popular Bellman equation. Finally, see what you should consider before deploying a trained policy, and overall challenges and drawbacks associated with this technique.