## Object-Centric Learning with Slot Attention

Video and paper.

tldr. but the routing mechanism here seems to be quite similar with the capsule one. The authors emphasize that a slot should learn not just one type of object. It seems that the main trick is to first route image feature to slots. Slots will then be train to fit not just one type of objects like capsule network.

## Set distribution networks

A work (video, pdf) from Apple to generate set with energy based model.

## Big Self-Supervised Models are Strong Semi-Supervised Learners

1. Self-supervised pretraining
2. Supervised finetuning
3. Distillation: train a student to learn the output of the teacher rather than the true label.

It seems to have a rather counterintuitive conclusion. Labeled data do not always help. Or too many labeled data used during training does not help.

## Quaternion explanation

A wonderful explanation of why $q p q^{-1}$ is rotation of $p$, where $q$ is most easily interpreted with the complex-number like structure of $\cos(\theta/2) + \sin(\theta/2)(a i + b j + c k)$ with $a i + bj +ck$ normalized (i.e., $a^2+b^2+c^2=1$). Note that $q^{-1}$ is then just $latex \cos(\theta/2) - \sin(\theta/2)(a i + b j + c k)$ (since $(a i + b j + c k)^2=-a^2 - b^2 -c^2 = -1$ just like the imaginary $i$ for complex number).

Basically quarternion is 4D stereographic projection back to 3D space. When multiply by a quarternion $\cos(\alpha) + \sin(\alpha) {\bf v}$ from the left, points parallel to $\bf v$ will be translate along $\bf v$ and points penpendicular to $\bf v$ will rotate around $\bf v$ (following the right hand rule). Similarly, when multiplying from the right by the same quarternion, points parallel to $\bf v$ will be translated in the same direction, but points perpendicular to $\bf v$ will be rotated in the opposite direction (following left hand rule). So if we multiple $q$ and $q^{-1}$ from both left and right hand sides, the translation effect will be exactly cancelled, and the the rotation will be doubled. That is why the angle in $q$ should be halved.

## Thousand Brains Theory

The conversation of Jeff Hawkins and Lex Fridman was very interesting. More info here.

## The New Skinner Box: Web and Mobile Analytics

This is old but is more relevant than ever. Everyone should read this.

## Population-based search

A great discussion on population-based search. Especially how it connects with goal switching and multi-task learning.