Personal Blogs
• Andrej Karpathy's blog: Probably the most didactic videos on practical use of deep learning on his YouTube channel.
• Luis Moneda's blog: Super interesting posts about out-of-distribution generalization and the best texts I've ever seen so far on career/management in data science on his DS leadership material.
• Lilian Weng's blog: In-depth articles on deep learning, presenting complex concepts with clarity and detailed explanations.
• Max Halford's blog: Very interesting posts about various aspects of data science, from database queries to using online models with streaming data.
• Matheus Facure's blog: The causal inference posts and the basic tutorials are really didatic and useful.
• Alessandro Takeshi's blog: Physicists love to solve difficult equations. Alessandro is a physicist. Jokes aside, although the posts are not trivial, they are much more digestible than the places where the ideas come from!
• Guilherme Marmerola's blog: A great place to discover different and elegant approaches. Any visual resemblance to this blog is purely coincidental!
• Caio Almeida's blog: Despite Caio being a physicist (not everyone is perfect), he is one of the most intelligent people I've had the opportunity to interact with.
• Matheus Jorge's blog: Alexandr.IA is an didatic resource for those who are having their first contact with some topic.
• Raphael Tamaki's blog: Very interesting posts about predicting customer lifetime value.
• Juliano Garcia's blog: Whenever I need to take averages of circular quantities (like the time of the day), I visit this blog.
• Pedro Tabacof's blog: General discussions around machine learning.
• Chip Huyen's blog: Lots of interesting material on machine learning in production.
• Alex Smola's blog: Despite him not updating his blog for a while, the content is still very rich, and I have a special affection for his post on covariate shift.
Industry Blogs
• Fast Forward Labs: Research reports and prototypes from Cloudera team.
• Distill: didactic and complete articles on Machine Learning.
Books
• A Course in Machine Learning: One of the best references for a first approach to Machine Learning techniques! Hal Daumé III manages to order the subjects in a logical and interesting way and provokes you to make the right reflections. After reading part of it, you can study the algorithms it doesn't cover with a deep understanding of what's going on. It is a modern book that is concerned with discussing relevant topics such as Fairness.
• Foundations of Machine Learning: If you want to delve into the details of Learning Theory, I highly recommend this book. Mohri discusses in depth the learning objectives, defining clearly and mathematically what we mean when we say that the algorithm "learned" through PAC-learning. It clearly motivates the minimization of empirical risk by rigorously defining the notions of complexity (with Rademacher's and VC-dimension). It's a very theoretical book, but it immerses you in the philosophy of Occam's razor.
Courses
• Learning from Data: Yaser Abu-Mostafa tricks you by translating complex subjects in a simple and didactic way. It's a great introduction to the same notions presented in Mohri's book, but in a visual way with lots of cool examples. It's a very cool deepening for those who already know the basics of Machine Learning and want to see the theory that guarantees learning a little more rigorously.