“Universal adversarial perturbations” seems most dramatic ML result in years; if so, not getting deserved attention https://arxiv.org/pdf/1610.08401v1.pdf …
-
-
The form of the universal adversarial perturbation is consistent with this hypothesis. It subtly screws up texture/color info:pic.twitter.com/tv3tRPp1UF
-
(In this picture, from the paper, the intensity of the perturbation is vastly magnified to show its form; it's actually undetectable to eye)
- 5 more replies
New conversation -
-
-
Convolutions are obviously going to be good at picking up textures (if you have enough of them). Hard problem in vision always was…
-
3D shapes, when projected in 2D image, are entirely different depending on rotation. Worse when object is flexible or articulated.
- 4 more replies
New conversation -
-
-
This Tweet is unavailable.
-
I vaguely think I remember that :) Has been 25 years since I was reading visual neurophys & psychophys
End of conversation
-
-
-
Likely. I'm of the opinion that classifiers need to be presented with multiple rotations and low-pass filtered versions and 1/n
-
if they disagree, it's a sign of a low confidence match.
End of conversation
New conversation -
-
This Tweet is unavailable.
-
-
I think separation of images into shaped surfaces is a fixed function in perception
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.