Absolutely agree most of the neural net worod fervently believes symbolist tradition took a wrong turn... but their argument rarely goes deeper than "it didn't work when we tried it" - which is a weak argument that was wrongly applied to deep learning in 2010. Very very weakhttps://twitter.com/intuitmachine/status/1200799872745652224 …
-
I would argue that the neural net world should study cybernetics and complexity science first before they bother with the symbolic tradition. Yes, I agree that the neural net community is missing something. I disagree though that it is the past knowledge in symbolic computing.
you are welcome to make the argument for cybernetics (which may have real value) but you have not to my knowledge given a substantive and compelling argument against including symbol-manipulation as part of the solution.
-
-
Only if what you mean by symbol-manipulation is equivalent to understanding causal models. Animals understand how to manipulate objects and this is equivalent to framing it as an 'algebra over symbols'. I prefer to avoid using this kind of abstract language.
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.