Welcome to Software-Over-Beer's course on Deep Learning!

• See below for status report

The What

This course will cover most of what there is to know about deep learning. We will start with the basics of a simple feed forward neural network, discussing forward and backward propagation. From there, we will dive into more complex neural networks.

The course will be divided into units, each (ideally) being self-contained. Each unit will cover the material from a theoretical point of view as well as a practical point of view. For the practical portions of the course, I will be trying to create hands-on exercises for you to attempt. Hopefully, this will help you better learn the material.

I should also note that the exercises will be available in BOTH PyTorch and Tensorflow 2.0. Since we are building from the ground up, I think this will offer you a good opportunity to explore both libraries and see which one you prefer (but it certainly doesn't hurt to be familiar with both).

Important: this course is not going to reinvent the wheel. If I think there are other resources out there that do a good job of teaching the material, I am going to direct you there. This course is going to tie in material from around the web, and supplement it with my own videos and exercises.

The Who

I'm a professional software engineer currently working at Amazon. I studied computer science at the University of Michigan, earning my bachelor's degree with a course-load that concentrated on machine learning and artificial intelligence.

With that said, a professor once told me that "you don't truly understand material unless you are able to adequately teach it to your peers." I think there's truth in that, hence this series.

If you have any questions, feel free to reach me at [email protected].

Final Notes

If you're looking for a way to quickly learn the ins and outs of deep learning, this isn't the place for you. I'm not sure there is such a place, to be frank. A true understanding of deep learning takes time and some hard work.

I believe if you put the time into this course, do the exercises, and ask questions, it will be worth your while.

Disclaimer: Part of the fun for me in making this course will be to teach it over a unique bottle of beer in each video. I will probably spend a few seconds kicking each video off with an introduction of that video's material and beer. If that does not interest you, just skip the first 20 seconds.

Feedback on course material is always welcome and appreciated (as are recommendations for beer)!

Status Report

05/30/2019 Created repository for this course.

Short term goals

• Complete section on prerequisites
• Complete section on setting up
• Complete Unit 1: An introduction to neural networks
• Complete section on the Universal Approximation Theorem

Long term goals

• Complete Unit 2: An introduction to Convolutional Neural Networks
• Lay out what other units will contain (RNNs, LSTMs, autoencoders, Reinforcement learning, etc)
• Create a web page to supplement this site (main purpose for now would be a comment board)