That's a simple statement of fact :-). It is actually quite narrowly drawn.
yes @yelcun connectionist imperialism exists, just as symbolic imperialism did earlier. a sample: “Take any old problem, where you have to predict something, and you have a lot of data, and deep learning is probably going to make it work better than the existing techniques."https://twitter.com/ylecun/status/1065648782736572422 …
-
-
-
On the other hand, the Physical Symbol System Hypothesis of Newell & Simon makes extraordinarily broad claims. I think N&S should be congratulated for being bold; it encourages us to find the weaknesses in their approach.
End of conversation
New conversation -
-
-
It's hilarious to watch two symbolic AI practitioners argue over imperialism, as if deep learning was somehow non-symbolic. What do you people think a label is if not a symbol? What is a deep neural net if not an expert system plagued with the same brittleness and other flaws.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What is the symbolic approach to generative models like style transfer GANs? What is the approach for RL problems? Or are you just talking about classification/regression problems?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.