The Hands-on Reinforcement Learning Course

From Zero to HERO

Out of intense complexities, intense simplicities emerge. – Winston Churchill

Contents

• Welcome to the course
• Lectures
• Wanna contribute?
• Let’s connect!

Welcome to the Course

Welcome to my step-by-step hands-on course that will take you from basic reinforcement learning to cutting-edge deep RL. We will start with a short introduction to what RL is, what it is used for, and how the landscape of current RL algorithms looks like.

In each following chapter, we will solve a different problem, with increasing difficulty:

• Easy
• Medium
• Hard

Ultimately, the most complex RL problems involve a mixture of reinforcement learning algorithms, optimizations, and deep learning techniques. You do not need to know deep learning (DL) to follow along with this course. I will provide you with enough context to become familiar with DL philosophy and understand how it becomes a crucial ingredient in modern reinforcement learning.

Lectures

• Introduction to Reinforcement Learning
• Q-learning to drive a taxi
• SARSA to beat gravity
• Parametric Q learning to keep the balance
• Policy gradients to land on the Moon

Wanna Contribute?

There are two things you can do to contribute to this course:

• Spread the word and share it on Twitter, LinkedIn.
• Open a pull request to fix a bug or improve code readability.

Thanks

Special thanks to all the students who contributed with valuable feedback and pull requests:

• Neria Uzan
• Anthony Lapadula
• Petar Sekulić

Let’s Connect!

Subscribe for FREE to the Real-World ML newsletter. Follow me on Twitter and LinkedIn.

Troubleshooting

If you encounter any issues during your journey through reinforcement learning, consider the following troubleshooting steps:

• Ensure that you have the correct version of Python and any necessary libraries installed.
• Revisit the lecture materials if a specific concept is unclear.
• Join community forums for additional support.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.

Code Explanation Analogy

Think of reinforcement learning (RL) as training a dog to fetch a stick. Each time the dog successfully retrieves the stick (the desired action), you reward it with a treat (positive reinforcement). However, if the dog ignores the stick and runs after a squirrel instead (an unintended action), it misses out on the treat (negative reinforcement). Over time, the dog learns to associate fetching the stick with rewards, gradually increasing its skills – this mirrors the process of RL where algorithms learn to optimize their actions based on feedback.