Thank you Gary -- the first author's handle is @JoaoLoula, and the wrong Brenden Lake got tagged (I'm @LakeBrenden)
More trouble in deep learning paradise, this time in generalizing language (exactly as anticipated in The Algebraic Mind).
Nice work by João Loula @BrendenLake and Marco Baroni keeping the seq2seq world honest. https://arxiv.org/abs/1807.07545
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
To my surprise, meaning is never a matter of curiosity in deep learning. Does a network know what does it mean to run vs. run slowly?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Everybody knows deep learning is not the answer. I and others have known it long before it became a household term. But you're one of several loud but hopelessly deluded voices who are promoting the nonsense that adding symbolic AI to deep neural nets is the answer. Pathetic.
-
BTW, I asked symbolic AI causality guru,
@yudapearl, the following question: "Unlike neural nets, the brain can instantly see a complex object it has never seen before. A DNN requires a prior representation. How does your work solve this problem?" Pearl had no answer. Do you? - 1 more reply
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.