Explore a code-driven introduction to Reinforcement Learning (RL) in this 33-minute conference talk from GOTO 2020. Delve into the fundamentals of RL, including the Markov decision process (MDP) framework, and learn how to build an RL algorithm to solve it. Follow along as the speaker demonstrates creating a "GridWorld" simulation from scratch, derives a simple RL algorithm to solve the simulation, and provides actionable steps for applying RL in industry. Gain hands-on experience with a provided Jupyter notebook and discover how RL, building upon machine learning, has the potential to automate strategic-level thinking in various fields. Perfect for beginners, this presentation covers key concepts, practical coding examples, and future applications of this cutting-edge AI paradigm.
A Code-Driven Introduction to Reinforcement Learning
Overview
Syllabus
Intro
Agenda
What is Reinforcement Learning RL?
Coding the MDP
Coding the RL solution
Next steps
Outro
Taught by
GOTO Conferences
