When the software world wants to build stuff that works, it still mostly uses stuff that innately incorporates symbol-manipulation. Can someone please show me an operating system, a word processor, a spreadsheet or scaleable DB that is built with neural networks and no symbols?https://twitter.com/tyrell_turing/status/1003313227852181504 …
-
Point is, this discussion seems to be shifting the innateness goalposts. NNs can learn symbol-like representations and abstractions, but the cited example domains don't use them because we want bespoke solutions rather than general AI.
What do you have in mind re NN’s and symbol-like abstractions, and what is innate and what is learned in the systems that you have in mind?
-
-
1/ I'm talking broadly about the ability to learn abstractions in different application domains (eg. hierarchical visual feature representations, high-level language structure, etc). But that's orthogonal to my main point, ...
-
2/... which is that "build stuff that works" is a different brief to "investigate whether solutions can emerge without hard coded machinery". From that perspective, the fact that people use hardwired solutions to the former doesn't really argue for innateness ...
- 1 more reply
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.