We are releasing an open dataset for theorem proving, HolStep: http://openreview.net/forum?id=ryuxYmvel … - can you beat our 83% accuracy baseline?
-
.
@fchollet sure, I'll play. 85%, took me about an hour. (totally possible I did something wrong in preprocessing though!)pic.twitter.com/RpLqmHiHPP
Cool! Our baseline is also just doing n-gram pattern matching, in essence (as we note in the paper).
-
-
Now it would be interesting to see if significantly different techniques can yield an accuracy much higher than 85%.
-
kind of disappointing context didn't help. I didn't try :/
End of conversation
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.