As someone who enjoyed your paper a lot, 3 things stood out:
1. The citation to @dawnsongtweets Berkely systems paper seemed ill informed. They do not cast doubt on Deep Learning at all.
2. The omission of Markov Logic Nets in your symbolic section was prominent
-
-
- 5 more replies
New conversation -
-
-
I liked the paper very much. One thing stood out to me as missing from your section on "insight from cognitive and developmental psychology": the ability to make analogies. IMO this ability is key to learning without big data (and "analogy" is not simply "transfer learning").
- 1 more reply
New conversation -
-
-
The operative word is "substantive". Not ad hominem attacks. So far, the only
@tdietterich made some valid pointsThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I posted an even more detailed rebuttal to each of your 10 criticisms of Deep Learning. https://medium.com/intuitionmachine/intuition-machines-versus-algebraic-minds-fad052b46ad5 … Enjoy!
#deeplearning#ai#mlThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Given that "deep learning" can (apparently) mean any big model that contains something differentiable, how, when history is written, will we be able to say definitively say whether you were right or wrong? In other words, what are specific bounds on your def. of deep learning?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Good observation. Actually, I suspect that there isn't enough funds to go around with AI research and that's why you have this kind of criticism against DL. I Ben Goertzel had to go do an ICO to get funding, that tells you that traditional funding has all dried up.
End of conversation
New conversation -
-
Thanks for the great paper! I wrote up some thoughts in the hope that they add to the discussion.https://medium.com/@christiankaiser_57972/thoughts-on-gary-marcus-critique-of-deep-learning-89146e63c3b3 …
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.