. @fchollet What do you think is a good direction for replacements?
I would bet against current backprop-based techniques in the long run.https://twitter.com/michael_nielsen/status/669978274701942785 …
-
-
-
@michael_nielsen Information geometry / topology will give us the theoretical foundation for building "optimal" learning algorithms. - Show replies
New conversation -
-
-
@fchollet@michael_nielsen better study of bio might also give fruitful results: our brains don't do backpropThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@robertsdionne@fchollet@michael_nielsen this is great as well: https://www.youtube.com/watch?v=Mcal5Cy7r4E … the heat-propagation ideas are fascinating.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@robertsdionne@fchollet@michael_nielsen A*, the comeback? ;-)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet@michael_nielsen I would bet on quantum algorithms of current methods.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.