Learning PCA? SVM? K-Means Clustering? Ask yourself these questions for a thorough understanding of these algorithm.
An (inexhaustive) list of questions to ask when learning about a machine learning algorithms. (WIP)

Variational AutoEncoders are hugely successful generative and disentangling models. IMHO, Variational AutoEncoders are more intuitive than AutoEncoders themselves. Don't believe me?
A semi-pedantic derivation of the VAE framework.

The ability of Neural Networks to generalize on unseen data has been a topic of debate for quite some time now. Check out the paper that won the "Best Paper Award" at ICLR 2017!
An analysis of the paper - Understanding Deep Learning Requires Rethinking Generalization by Zhang et al

The ability to read through thousands of papers within a short amount of time, is a super power we all wished we had. Thanks to Machine Learning it's now possible!
Nanomaterial Synthesis Insights from Machine Learning of Scientific Articles by Extracting, Structuring, and Visualizing Knowledge by Hiszpanski et al