Delta Academy Tutorials
Delta Academy Tutorials

Delta Academy Tutorials

Intro to Reinforcement Learning for Coders

Preface: What is this and who is it for?

The below are the tutorials and Python exercises for the Intro to Reinforcement Learning course for coders.

We released the tutorials and exercises for free because we felt a practical & accessible reinforcement learning course with coding exercises isn’t widely available.

The focus of the tutorials is practical. They contain interactive coding exercises in Python. Each exercise has tests to check your solutions. The tutorials cover the necessary theory without delving into detailed mathematical proofs. However, we recommend you are familiar with mathematical notation typical to an undergraduate-level technical course. This is because key equations and theorems are best stated mathematically.

The course also assumes a solid grasp of the basics of Python. It’s ideal for software professionals, product managers and interested hobbyists who want to dive into Reinforcement Learning.

These tutorials were built to accompany Delta Academy’s 4-week “Intro to Reinforcement Learning” cohort-based course. Every week, you apply what you’ve learned in an AI-building competition. In teams of two, you build an AI which competes with the rest of the cohort in well-known games - such as Connect-4, Poker and Pong. The course also includes weekly Office Hours with an expert and a cohort Slack workspace.

If you find issues, or errors or have suggestions, please reach out to

Intro to Reinforcement Learning for Coders

Week 1 - Reinforcement Learning Fundamentals


1 - Motivation, States, Actions and Rewards
2 - Return, Value Functions & Bellman Equations
3 - Learning from Experience, TD-Learning, epsilon-Greedy

Week 2 - Function Approximation


4 - Generalised Policy Iteration
5 - Curse of Dimensionality, Function Approximation & Parameters
6 - Feature Vectors

Week 3 - Deep Neural Networks


7 - Intro to Neural Networks
8 - PyTorch
9 - Training in PyTorch

Week 4 - Deep Q-Networks


10 - Model-free control & Q-Learning
11 - Deep Q-Networks
12 - Double DQN


This toggle block is hidden on your Super site

Content Database ↓