One reason the puzzle of intelligence is so difficult is that it has so many pieces. Perception, control, abstraction, curiosity, reasoning.
And many more. Hard to blend these concepts into a single coherent framework that would explain biological minds & allow to create new ones.
-
-
The framework that blends all of these together is called meta-RL.
-
no, very far from it.
- Show replies
New conversation -
-
-
especially as the framework does processing, reasoning, AND storage with the same architecture!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
that's basically the aim of the book The Measure of All Minds. Recommend if you haven't read.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
another implicit assumption is that these concepts adequately explain the entirety of intelligence.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
maybe a single coherent framework will never achieve intelligence if
@danieldennett is right in his phenomenal book: from bacteriaThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Maybe focus should be less on combining any of these "separate" parts, which are products of our (possibly misleading) intuition.
-
Engineering a model of intelligence should yield any of these qualitative parts automatically as they must be related by necessity
End of conversation
New conversation -
-
-
deep learning is great, but biological minds also do protein folding which is o(2^n) in software.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
are these really necessary? or they only how we apprehend intelligence? Maybe the actual intelligence doesn't formalize these...
-
like particle swarm algo which is quite simple but match very complex biological swarm behaviors.
End of conversation
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.