What are the best deep learning books to read?
A fun read is Pedro Domingos’ “The Master Algorithm”. The book talks about the quest of researchers to build a truly artificial general intelligence. While the second part of the book is less interesting, the idea conveyed in the book definitely get hyped up about the promises on AI. A more theoretical book, and a masterpiece, is Kevin Murphy’s “Machine Learning: A probabilistic approach”. If you’re looking to write a thesis on Data Science, this book shouldn’t be far from your desk.
As a voracious reader, I enjoy reading both technical and non-technical books. My 2c on a few good books:
- (1) Sebastian Raschka’s “Python Machine Learning (2nd ed)” is an excellent, practical overview of traditional ML.
- (2) Aurelion Geron’s “Hands on Machine Learning with Scikit-Learn and TensorFlow” is a superb guide on ML and DL with TensorFlow. I would recommend it to a practitioner
- (3) The book “Prediction Machines - the Simple Economics of AI” (by Ajay Agrawal et al) is a great big picture book on the practical applications and implications of large scale AI
I am reading the Deep Learning book by Goodfellow et al. I found this to be the best book to understand the concepts related to Deep Learning intuitively. The authors have explained it in the most beautiful way one can. Here are the links. 🙂
Do videos count as "books"?
I watch much more videos than read books, like those by Andrew Ng (of course!!!), many others. Books are out of date mostly and slower to input to your brain. Seriously, I don't read books any more just about. All the best universities and other producers have great DL and ML content online in videos now.
The best videos then:
- Too many to list. If you are intro level to machine learning, and already completed Andrew's MOOCs, then I will recommend Statistical Learning MOOC at lagunita.stanford.com by Tibshirani+. Also, Prof Rafael Irizarry of Harvard U's data science certificate sequence of courses at EDX website, which is basically data science for modern biology. If you are expert level or intermediate, then I will recommend watching all the latest current industry talks and shows in your areas of particular interest on video at youtube produced by the many working practitioners of data science at lots of different kinds of companies.