A Code-Driven Introduction to Reinforcement Learning

In this presentation I present a code-driven introduction to RL, where you will explore a fundamental framework called the Markov decision process (MDP) and learn how to build an RL algorithm to solve it.

REINFORCEMENT LEARNING
TALK

Notebook link

Abstract

Reinforcement learning (RL) is lined up to become the hottest new artificial intelligence paradigm in the next few years. Building upon machine learning, reinforcement learning has the potential to automate strategic-level thinking in industry.

In this presentation I present a code-driven introduction to RL, where you will explore a fundamental framework called the Markov decision process (MDP) and learn how to build an RL algorithm to solve it.

First I show you how to create a simple “GridWorld” simulation of the MDP, from the ground up, to help demonstrate why and how RL works. Then I derive a simple RL algorithm that’s capable of solving your simulation. Finally I will provide actionable next steps to show you how to take this learning and apply it to industry.

This presentation includes a Jupyter notebook that you can tinker with during the presentation. Full instructions will be provided. Although this presentation is suitable for beginners, you will benefit if you have some exposure to data science and machine learning.