You are hyping anti-hype. It's a really cool paper, but calling it 'devastating' isn't particularly helpful. For example, ResNets (which have been the standard for a few years!) seem a lot more robust.
devastating new research showing serious problems with robustness in deep nets by @beenwrekt; fits perfectly with the recent @bethgelab work, and my own critical appraisal of deep learning.https://twitter.com/maxalittle/status/1003650956716855296 …
-
-
-
you have data to show resnets are more robust here?
- 2 more replies
New conversation -
-
-
Just want to say that almost all of the work was by
@lschmidt3,@BeccaRoelofs, and@Vaishaal!Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Not a surprise at all. Just ask a PhD in Mathematics about what advanced calculus n-space topology reveals as implicit. PhD's in
#AI ignore math realities at their ongoing peril as illustrated by this conundrum described here.
-
Hello, can you give me a link to further information on what "advanced calculus n-space topology reveals as implicit" - I am curious what you are pointing out here ..
- 2 more replies
New conversation -
-
-
AI research stuck with whipping the dead horses namely MINST and CIFAR datasets. As long as there is no REAL LIFE robustness, who cares?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
This Tweet is unavailable.
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.