Major news re #deeplearning & its limits: Yoshua Bengio’s lab confirms a key conclusion of Marcus 2001 and 2018: deep learning isn’t efficient enough with data to cope with the compositional nature of language. https://export.arxiv.org/abs/1810.08272
-
I agree. Is
@GaryMarcus' claim go beyond "neural networks is not enough"? What then is he proposing? Is it 'symbolic machinery'? Is the claim that 'symbolic machinery' cannot be simulated by neural networks?
And yes i am proposing symbols and see new post on Medium, to clarify. (Link posted here a moment ago)
-
-
I will point out that a recent paper: https://arxiv.org/pdf/1810.10531.pdf … reveals that the semantic symbolic representation you seek are indeed already present in existing neural networks. Therefore, falsifying your theory. Comments?
-
I have seen people fit requirements to their design. It seems to be the case with AI as well. Its like a generic graph/mathematical data structure is been developed and we try to fit common sense logic to it. Need is to design a DS that represents common sense inherently. 1/2
- 4 more replies
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.