astonished that anyone knowledgeable could claim that neural nets are (obviously) an “abstraction of neural processing” when we don’t yet know how brains work. if you don’t know how Y works you can’t really speak with certainty about whether X is an abstraction of Y. Period.https://twitter.com/tyrell_turing/status/1200072223299657728 …
-
Oy. “Knowing how Y works” means having a theory of Y that explains something and hasn’t been contradicted by evidence, yet. “X is an abstraction of Y” just means a theory of Y based on a simpler theory X. How can you ever “know how Y works” without making abstractions?
symbol-manipulation has not been rejected by evidence yet, but most readers of this thread are "sure" the brain doesn't do it (despite abundant evidence from linguistics and psychology that suggests symbol-manipulation is part of what brain do)
-
-
But you probably can do symbol-manipulation with ANNs, why not. You'd just need a larger network and some fun architectures. It doesn't negate the fact that the brain is made of networks, and ANNs are mathematical abstractions that capture most interesting facts about networks.
-
would urge you to read my lengthy discussion of this in the Algebraic Mind, chapters 2 and 3
- 2 more replies
New conversation -
-
-
Not here to fight, but may I ask how symbol-manipulation is implemented in the brain (if not by a network of interconnected neurons)?
-
Don't know the answer to this, but (a) I made some suggestions in The Algebraic Mind, chapter 3, and (b) I also don't know how short-term memory (of the write-once, then immediately retrieve variety) is implemented in the brain, but that doesn't make me doubt its existence.
- 4 more replies
New conversation -
-
-
I can't speak for the rest of them. I think the brain does symbol manipulation and therefore subscribe to "implementational connectionism." But I'm trying to work out the implementations, so that we can actually use the power of both theories! Hope to show some progress soon.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Has anybody attempted symbol-manipulation for Sanskrit? ANNs seem a waste when the grammar is so clear:https://twitter.com/prem_k/status/1198481799644495873?s=20 …
When the grammar is so simple (few mathematical rules when applied to a few root words leads to a rich language) do we need#DL/#ANN to build#NLProc, either for#NLU/#NLG?
Admire the genius of Sanskrit. It's even got a tensor of its own! #AI Source: https://youtu.be/0emIewicwl0 pic.twitter.com/lTJl92HgbLThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
nor has the immortal soul Keep the faith!
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.