News for hackers. Stuff that matters.
Re:Every Model Learned by Gradient Descent Is Approximately ... by FRGabriel on Sunday December 06, 2020 @03:10PM (#25326245)
So at the end, it rephrased a statement from "Neural Tangent Kernel: Convergence and Generalization in Neural Networks" [https://arxiv.org/abs/1806.07572], besides in a way which is kind of miss-leading.

The assertion is known by the community at least since 2018, if not even well before.

I find this article and the buzz around a little awkward.

Re:Every Model Learned by Gradient Descent Is Approximately ... by QuesnayJr on Sunday December 06, 2020 @03:49PM (#25326486)
To be fair, the article you cite is from 2018, and the OP is from 2012 (if the arXiv dates are right).
Re:Every Model Learned by Gradient Descent Is Approximately ... by ajtulloch on Sunday December 06, 2020 @03:54PM (#25326533)
The OP article was submitted on Mon, 30 Nov 2020 23:02:47 - the arXiv identifier 2012.xyz means it was published in the 12th month of 2020, not 2012.
Re:Every Model Learned by Gradient Descent Is Approximately ... by QuesnayJr on Sunday December 06, 2020 @04:32PM (#25326861)
Oh God that's embarrassing. I really thought that's how arXiv identifiers worked, and have been under the wrong impression for some time.