AIDL Weekly Issue 36 - Capsules, Capsules and Capsules


Thoughts From Your Humble Curators

We were out last week. The hottest news this week is all about Prof. Hinton's capsule models!

Prof. Hinton and students just released an arxiv paper on how the idea of capsule can be used and specifically how it can outperform MNIST as well as its more difficult cousins, affMNIST and MultiMNIST, which had distorted MNIST with affined transform and heavy overlapping. So we dedicate this issue on capsules. We provide our own analysis in the thesis/paper review section, highlight a popular link from Wired, and cover the latest developments. So far, we already know of two implementations which attempt to repeat the result.

Other than that, check out other interesting links such as ex-Google Brain Resident David Ha's work on evolution strategy, and our piece on "Unsupervised Machine Translation Using Monolingual Corpora".

Join our community for real-time discussions with this iOS app here:

As always, if you like our newsletter, feel free to subscribe/share the letter to your colleagues.

We will host our own show at the AI World events: "Attack of the AI Startups". If you live in the New England area, feel free to join!

Artificial Intelligence and Deep Learning Weekly


Blog Posts

Open Source

Paper/Thesis Review


Leave a Reply

Your email address will not be published. Required fields are marked *