Asking to learn. I can understand low utility. Isn't it interesting to explore the boundaries of DL in research?
#nips review for modeling #math functions using #lstm: already have software for that. no need for #deeplearning #cluelessNIPSreviewers
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
that wasn't me (I didn't review for NIPS) & i wouldn't reject for that, but I'm curious your motivation? my bias is to learn to use libs :/
-
if I understand her point correctly: teaching NN to perform mathematical operations is a genuinely cutting edge area of research.
- Show replies
New conversation -
-
-
Boo! Lame.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
WTF, you'll never get an LSTM to do pure math to pro research level, beyond maybe simple theorem checking? Have u ever done real pure math?
pic.twitter.com/Cv7BveCJB2
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Our reviewer: why performance has no gains over "competitor". "competitor" was the MLE of the generative model (so cannot be beat).
-
So our method learned the MLE optimal solution without access to the generative model.
#CluelessNIPSreviewers - Show replies
New conversation -
-
-
my q is: if ur goal is practical, why force NN to relearn things machines are already GREAT at vs teaching them to learn to use libraries?
- Show replies
New conversation -
-
-
In our NIPS review, reviewer makes multiple interesting comments of the form "O(log(x))!=O(log(x^2))"
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.