ML systems/libraries treat their input as a table (or "data frame") but that's an abdication of the hard problems which all involve time.
-
-
yeah, if you let them slice it up into lumber before you get it there's just so much more waste.
- End of conversation
New conversation -
-
-
imo not having to manually tabulate rollup feats from a stream is the biggest potential gain from "representation learning" / NNs
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
that's true for online learning algs; but most of ML boils down to a system of linear eqs and hence the matrix of observations :)
-
at least historically. I'm sure there will be renewed interest in online algs and never-ending learning because of logs
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.