So I have not uploaded a new post in over 6 months, I figured a would upload an update on what I’ve been up to and what to expect in the future.
Nerd Alert: You probably won’t be familiar with the majority of what I’m talking about unless you have some knowledge of the ML field.
The study of flow and manipulation of information in systems has always been a passion of mine, ever since I worked through Andrew Ng’s landmark course when I first got to Olin. But, I never had the time or coding/math/stats experience to do a deep dive pun intended until spring 2018. Starting then, I began my Learning Deep Dive with NNOSS. Some friends and I wanted to deepen our knowledge of Neural Networks and subsequently created our Neural Network Olin Self Study. As we could not find a decent concept map online, we decided to create our (shown below) with links to reports with more information.
If you want access to the map, just email me and I can send you the information.
After spending most of the semester researching and expanding this concept map, we moved onto projects. I chose the Deep Convolutional GAN (on which I will have a post later).
After NNOSS ended, my summer began and I continued my dive by building a Deep Learning Machine, which you can read about here.
For the final stage of my learning deep dive I spent the a large amount of my summer coding and reading any Deep Learning literature I could get my hands on including (but not limited to):
- Introduction to Statistical Learning
- Deep Learning
- 100 Influential ML papers from 2012 – 2016
- Good Ole Arxiv
- This week in Machine Learning
- My twitter
As far as tinkering goes, I put all my work on the NNOSS Github Page or my Github. Be forewarned, most of the work I’ve done during Summer 18 has been for personal growth and not presentation, so most these repos are pretty rough around the edges.
As I didn’t spend 6 months working hard on this for nothing, expect more DL projects in the future., and as always if you have any questions, comments, concerns feel free to reach out.