I recently went on an internet spree to get myself versed with fundamentals of deep learning and more importantly intuition of the same. It has been an amazing journey so far, and thanks to all the amazing teachers on the web, I have a firm ground on which I can start studying this topic further. Here are the most interesting links I found to understand intuition (not your typical moocs and courses and articles which are cataloged just about everywhere)If you struggle with math like I do, first go binge watch the couple of series by Grant Sanderson I have linked here:
Videos you shouldn't miss
Blogs and Articles you could subscribe to:
If you understand by visualizing, Chris Olah's blog is a great resource
Andrej Karpathy ( Imagenet > OpenAI > Tesla). Probably the leading authority on computer vision today
Bookmark this cheatsheet of cheatsheets. It will be a great reference:
Nice articles that go into the details of Backpropogation if you want to implement your own:
For RNN's and LSTMs read these two blogs specifically:
Word2Vec: If you want to understand in depth the word2vec such that you can implement it in code and understand full intuition of it, please look at Ajit Rajasekharan awesome explanation here: https://www.quora.com/How-does-word2vec-work-Can-someone-walk-through-a-specific-example/answer/Ajit-Rajasekharan?srid=32voX
Latent Dirchilet Allocation or LDA is a popular approach to build topic distributions of documents based on a topic model derived from a universe of knowledge. Below is a very nice explanation of the concepts:
https://www.youtube.com/watch?v=3mHy4OSyRf0
Bayesian Inference: The more I learn about machine learning, the more i realize the importance of Bayesian theorem in almost everything we are doing in ML. Thus I consider it is very critical to understand Bayesian theorem well to understand machine learning. Here are couple of videos. Be patient, you will get it.
https://www.youtube.com/watch?v=BcvLAw-JRss
https://www.youtube.com/watch?v=5NMxiOGL39M