@fchollet have you seen this paper? They claim 99% on cifar
https://arxiv.org/abs/1610.02306
Imo, i think it's beyond the bayes rate.
it sounds like it's better than the noise rate in the labels. I'd approach this with a dose of skepticism.
-
-
yeah, that's what i was thinking. You did an amazing job of exposing the sarm paper, so i was asking your opinion :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.