News for hackers. Stuff that matters.

A comment on Every model learned by gradient descent is approximately a kernel machine (2020). Parent comment

Re:Every model learned by gradient descent is approximately ... by XenophileJKO on Sunday February 25, 2024 @02:06AM (#39498367)
The key isn't that they just "mix patterns". They learn to mix higher order abstractions, which is fundamentally different from "memorization".
[ Parent | on HN ]

There are no comments on this comment yet.