Correction: you are attempting to hijack an opinion that Hinton was arguing for in 1991, while you were still in a larval stage. Hinton's view on the topic hasn't changed: how to implement reasoning with neurons and make it compatible with gradient-based learning.
Indeed, once upon a time, Hinton was interested in the same sorts of unification that I believe to be so important.https://twitter.com/danbri/status/1075653592596377600 …
-
-
-
This is a total misrepresentation of Hinton. Hinton was explicitly, then, trying to represent symbols & neurons in a common framework; he explicitly doesn’t want that now. You falsely equate “reasoning with neurons” (which we all agree brains do) with answer to symbol question
- 9 more replies
New conversation -
-
-
Origins of connectionist symbol processing: Ising & Lenz were the pioneers in Physics to first invent such connectionist symbol processing for a natural phenomenon. Ising, E. Z. Physik (1925) 31: 253.https://doi.org/10.1007/BF02980577 …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Aren’t computers symbolic all the way up and down?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Unfortunately for deep learning godfathers and symbolic AI crackpots, they will have no role whatsoever to play in the coming of AGI. The cluelessness of mainstream AI is so blatant, it should be a scientific scandal.https://twitter.com/RebelScience/status/1075448598458916864 …
- 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.