Deep Learning with H2O and R In my previous TNO4U talk I gave an introduction about how I addressed the classification problem for autonomous driving using fuzzy logic based insights. I also gave a very concise introduction on deep learning. In this talk I want to go more into the details of deep learning - what is it - and why people think it is so important. Due to the duration of the talk I will not go through the complete history of Artificial Intelligence from the perceptron, via the Hopfield net, towards modern Restricted Bolzmann Machines and Convoluted Neural Networks. Nor get philosophical and do a Gödel, Escher, Bach exposé. I will just give some basic theoretical considerations and demonstrate how one easy it is to get results with deep learning using – open source- tools like R and H2O. You can install these for free on any computer, Windows, Linux or Mac. R is of course the computer language of choice for data science, H2O is an easy to use interface between R and Big Data (like Spark). During the talk we will do some small workshop style examples. Handwriting recognition with a Restricted Bolzmann Machine, analyze heartbeats with machine learning and do a little predictive modelling on an industrial process. Are this the heartbeats of a healthy person? Let’s ask our algorithm (The computer has seen more heartbeats than any living doctor)