News for hackers. Stuff that matters.

A comment on Why do tree-based models still outperform deep learning on tabular data?. Parent comment

Re:Why do tree-based models still outperform deep learning o... by eru on Thursday August 04, 2022 @12:30AM (#32339378)
Lots of linear functions with \x -> max(0, x) thrown in between.

(That's literally what neural nets with relu as activation unit do.)

[ Parent | on HN ]

There are no comments on this comment yet.