It's one of hundreds of experimental results showing your claims are false. My students post blogs every week with high accuracy results on small datasets. I know you know that high accuracy on small data is possible so I don't understand why you're saying it's not
Toy data. Urge you to read “limitations” section of that JAMA paper, re other pathologies, rare events that were excluded, etc.https://twitter.com/jeremyphoward/status/941387676376113153 …
-
-
-
Being a deep learning skeptic is very hipster cool but it requires constantly moving the bar.
- 1 more reply
New conversation -
-
-
You can't call that paper "toy data". That is ridiculous. It is great data, for one task. Complaining that it ignores other pathologies completely misses the point.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
It's fine, as long as you're informed and being honest about experimental results. Claiming a technique can't do something that it does all the time isn't science
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You must be so proud to be "very hipster cool"!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I know of literally no deep learning researcher who claims the first conclusion. This is a straw man. I thought the result was cool, but entirely unsurprising
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.