Typical crowd-sourcing: collect k labels per example & majority vote. New work with Ashish Khetan & @AnimaAnandkumar provides theoretical and empirical evidence that when data is plentiful, it's better to label once and move on! https://arxiv.org/abs/1712.04577
-
-
interested in hearing about this also. I can imagine examples in medical imaging
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.