Introduction to reinforcement learning with Atari Games by Daniel Cook

March 2016, AlphaGo beats Lee Sedol. Reinforcement learning, so hot right now! Baffled by the Bellman equation and gradient backpropagation, I had to sit through a Data Science degree to get to the bottom of it. Why does everyone prefer to explain things with maths and obscure symbols? This talk takes a step back and will run through the intuition and practical code examples of both reinforcement learning and convolutional neural nets so you’ll be able to create your own AI to win at Pong or whatever Atari game takes your fancy.


Supported by