Retweeting my own tweet, because I don’t want it get lost in the stack of discussions about less relevant issues. THIS is what the debate w @ylecun is really about: whether or not we need machinery for performing operations over variables, in addition to vectors and gradients.https://twitter.com/GaryMarcus/status/1071452576074280960 …
-
Symbolic operations can certainly be learned by differentiable frameworks. But by default they learn propositional rules and not first-order rules (with universally quantified variables). Neural networks suffer, by default from what McCarthy called the "propositional fixation".
Yes, I agree; this was more or less the main force of The Algebraic Mind, and why I like your work @LittleBimble (& @egrefen , inasmuch as you are trying to address it.
