This happens when you try to apply a linear method (SVD) to a non-linear problem, i.e., why deep learning is hot: http://danluu.com/linear-hammer
-
@danluu@astrieanna Not sure linearity is the right framing of the problem. Many non-SVD methods are linear in inputs.
@johnmyleswhite @astrieanna How would you recommend framing it? This is just something I play around with, not my day job :-).
-
-
@danluu@astrieanna I think you're getting at a point that we lack a good vocabulary for. I'd tend to talk about objective functions.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@danluu@astrieanna SVD minimizes squared error, which isn't always appropriate for the same reason that median better than mean sometimes.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@danluu@astrieanna Anyway, sounds like it wasn't the ideal place for you to work at. Good luck finding something more interesting.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.