“Universal adversarial perturbations” seems most dramatic ML result in years; if so, not getting deserved attention https://arxiv.org/pdf/1610.08401v1.pdf …
-
-
You might be able to cheat just by being good enough at image segmentation?
-
That’s certainly an important part at least! I was working on segmentation in early 90s but computers were too #%^*+ slow
End of conversation
New conversation -
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
wonder if such an perturbation exists for human vision? or maybe processing takes place at too high of a level for that
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
when papers have explored what the conv nodes "see", there is structure at higher levels, not just texture
-
but built up from Gabor-like filters, and these perturbations seem to confuse things at that scale
End of conversation
New conversation -
-
-
I've thought similar. Foreground/background segmentation, scale estimation, 1/n
- 2 more replies
New conversation -
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.