tahw

How could this possibly be faster than a linear classifier? In "Implementation" near Section 2 they claim that their implementation is better than a linear classifier but that seems like it couldn't possibly be true could it? A single floating point operation per parameter has gotta be faster than multiple branches, right?

This is not mathematically necessary at all. This just reads like astroturfing for an overpriced mic retailer.

Do you have a source on the orangutan grader? That sounds like an hilarious read.

Ad companies that specialize in re-targeting are going to have an absolute blast with this!

Its kinda poetic that this page is down right now.

Can anyone explain in layperson terms how this is different from regular old polarization?

They have no way to tell who is/isn't a citizen so just change your location to Frankfurt and voila, you're protected by GDPR!

Pay, Expected Weekly Hours (especially if it's salary), and Tech Stack are the most important to me.

A person that struggles with CSS and performs better when two other developers are telling them what to do should probably just not be in this industry at all. This sounds like a nightmare for any competent developer.

It isn't luck, it's money. Successful entrepreneurs aren't lucky or hard-working, they're just people who have enough cash to operate at a loss while they pursue an idea.

How the hell did they measure clock speed if the passage of time itself is different?

"Imagine you install analytics on a traditional content site. A few months later you update that site to be a single page application (SPA) without changing your analytics code. Then, a few months after that, you update your site to be a progressive web app (PWA) "

Who on earth would ever do this? The worst boss of all time?