@GaryMarcus Deep Learning must also be a "fad":https://twitter.com/PessimistsArc/status/1180157317058383872 …
-
It took several decades for connectionist ideas to return to acceptable scientific discourse: https://neurovenge.antonomase.fr/ .
and it may take several decades for symbolic ideas to return to centrality, but their current unpopularity doesn't mean that they aren't important. imagine the world if people had abandoned deep learning in 2010...
-
-
I think you conflate the value of symbolic approaches in computer science and its value in understanding human cognition. Symbols are useful only in cognition when it maps to human native capabilities. It's the difference between doing math using Roman and Arabic numerals.
-
this is just an empty assertion with no evidence
- 3 more replies
New conversation -
-
-
Symbolic AI is, of course, nonsense. Symbols are notoriously and inherently specific, the opposite of general. In fact, the reason that deep learning is not general is that it uses symbols, i.e., the labels. And it's still a rule-based expert system: IF pattern A THEN label X.
- 2 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.