[dontlikeyoueith]

> In the limit of training on a diverse dataset (ie as val loss continues to go down), it will converge on the process (whatever that means) or a process sufficiently robust.

This is just moving the goal posts from "learning the actual process" to "any process sufficiently robust"

[famouswaffles]

I didn't move anything because last i checked the term was Artificial Intelligence not Artificial exactly as a human does Intelligence