The Hands-on Reinforcement Learning Course
Reinforcement Learning (RL) is the kind of machine learning closest to how humans and animals learn. It offers us a path towards building general AI systems that can tackle the most complex problems we can think of.
1. Introduction to Reinforcement Learning
This first part covers the bare minimum concept and theory you need to embark on this journey, with practical examples and the first code snippet!
2. Q-learning to drive a taxi
Q learning is a classical RL algorithm published in the 90s. In this first lesson, we use tabular Q-learning to train a smart taxi driver. Ready to drive?
3. SARSA to beat gravity
The Mountain Car problem is an environment where gravity exists (what a surprise) and the goal is to help a poor car win the battle against it.
SARSA is a classical online algorithm that solve this problem like a charm.
4. Parametric Q-learning to keep the balance (1/3)
Parametric Q learning combines the strengths of classical Q-learning with modern optimization techniques from Supervised Machine Learning.
5. Deep Q learning to keep the balance (2/3)
Let’s replace the linear model from the previous lesson with a deep neural network. And kick-ass solve the Cart Pole environment.
6. Hyper-parameters in Deep RL
Hyperparameters in Deep RL are critical to training successful agents. In today’s lesson, we will learn how to find the ones that make you a happy Deep RL developer.
Deep Learning: Faster, Better, And Free In 3 Easy Steps
Tired of training deep learning models on your laptop, at the speed of… a turtle? 🐢 Not enthusiastic about buying an expensive GPU or optimizing cloud services bills? 💸 Wish there was a better way to do it?
Luckily for you, the answer to the last question is yes.
7. Policy Gradients to get to the Moon
Policy gradients are a family of powerful reinforcement learning algorithms that can solve complex control tasks. In today’s lesson, we will implement vanilla policy gradients from scratch and land on the Moon 🌗.
