I know how it feels you are sitting at the bank of the river and thinking whether to jump or not. It gets terrible when you don’t know how to swim. i was in a similar situation. It was couple of years ago when I was talking to my friends in IIT, some working in MSR(Microsoft Research) Bengaluru etc and I realized that I need to understand the field of deep learning then only I will have something to say in their conversation. Yes, getting into deep learning was not a career choice for me, it was peer-pressure. My friends like Asim Unmesh and Jeetu Raj were really helpful when I decided to pick up deep learning.
Now I won’t bore you with my story, let’s get to the exciting part, how I taught myself deep learning and machine learning? Excited?!
Disclaimer – The following list is about the resources which worked for me, I ain’t saying this is the only way to learn ML.
I will list down the resources for which I am thankful for. I really am glad that the creators left these resources for free online for folks like me.
- Introduction to Machine Learning (Andrew Ng, CS 229 Stanford) – I am thankful that I referred to this series because Prof Ng really takes us deep into Machine Learning intuitively and step by step. Make sure that you also read lecture notes.
- Introduction to Deep Learning (Andrej Karpathy, CS 231n Stanford) – Being super interested in Computer Vision, I wanted to learn more about this field. I wanted to know how deep learning is used in CV. Don’t forget to check out lecture notes.
- The Deep Learning Book – I picked up this book after I had finished more than half of CS229 lectures. This book also has some introduction to essential mathematics which are necessary for deep learning particularly.
- Introduction to Natural Language Processing (Christopher Manning, CS 224n Stanford) – For the folks interested in NLP, I’d recommend this lecture series. Although I still need to finish this series but I am finding it fun to learn from.
Great! Congratulations! if you have completed the course and made this far… I’d now try to provide some other reading resources which I refer.
- Sebastian Ruder’s newsletter – Please subscribe this awesome newsletter for latest updates from NLP domain.
- I keep a constant watch at ICML, CVPR, NeurIPS website to check at submitted and selected papers. This helps me to keep myself updated with the state of the art.
- Please subscribe to OpenAI, Google Brain and DeepMind’s blog, newsletter and twitter accounts.
Some more content –
- A survey of dimensionality reduction techniques – much of the data is highly redundant and can be efficiently brought down to a much smaller number of variables without a significant loss of information. The mathematical procedures making possible this reduction are called dimensionality reduction techniques and this paper surveys several techniques.
- Introduction to CUDA – A good short pdf which will give you a good understanding of CUDA. Also here is an amazing talk by Justin Lebar – “CppCon 2016: Bringing Clang and C++ to GPUs: An Open-Source, CUDA-Compatible GPU C++ Compiler” .
Want to get started with Reinforcement Learning? The following videos can be a good starting point –
- Introduction to Reinforcement Learning by Richard Sutton
- Also there is a great book available for free!
There is way more content which I have, please feel free to comment if you want anything specific.
Have any doubts? or want to have random chit chat about Deep Learning? then feel free to ping me up. Also please follow, subscribe and share this blog!
Happy Coding! 🙂