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 …
-
"They continue to develop the connectionist framework not because they are blind to its current limitations but because they aim to eventually bridge the gulf by building outwards -" [...]
yes, exactly, and a research program i fully approve of. and now compare that (“eventually bridge”) with the symbols-are-like-aether stance that he had adopted by 2015, https://sites.google.com/site/krr2015/home/schedule … & at recent G7 talk. the two stances can’t both be right.
-
-
"The ability to represent complex hierarchical structures efficiently and to apply structure sensitive operations to these representations seems to be a essential. Most connectionist researchers accept this," [...]
-
..."[Most connectionist researchers accept this], though they expect that this ability may be implemented in ways that have not been anticipated within the standard symbol processing tradition." Pretty clear, and consistent with modern efforts and attitudes.
- 3 more replies
New conversation -
-
-
https://twitter.com/santoroAI/status/1076161340681715714?s=19 … already said it - "cannot handle structured knowledge" can be taken as a statement on these artifacts' capabilities in the world (remembering our shopping lists, traffic laws) or a claim about how they work. Latter is looking increasingly aetherial...
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.