Here’s what I have been saying since 2001: “Neural networks have to be supplemented with the key components of symbol manipulation in order to capture the compositionality of language and thought.” Agree or disagree, @ylecun? At our debate you said “diagree”. Your stance now?https://twitter.com/ylecun/status/1065298686413078528 …
-
The only interesting question is "how?"
No, the interesting question is whether, and you have denied it, explicitly in print and conversation. Have you changed your mind?
-
-
Your statement is irrelevant because it is not falsifiable. The minute a learning machines will be capable of "symbolic" reasoning, you will declare victory and say "I told you so", regardless of the method. So, I repeat, the question is how.
-
Perhaps
@GaryMarcus is claiming that we will need to explicitly add "symbolic components", specifically designed to power symbolic reasoning in a targeted way; whereas@ylecun thinks it might be possible for symbolic reasoning to develop emergently in a more agnostic substrate? - 3 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.