@garymarcus Who says you need symbolic AI when Deep Learning can solve the 3-body problem? ;-) #deeplearning #ai #gofai https://arxiv.org/abs/1910.07291
-
For simulating physical systems (very important for science), your argument of human-like cognition in deep learning is not as relevant. Existing numerical methods consume inordinate amounts of computation and the unique intuition-based cognition is extremely valuable.
Not really sure what you are saying, and don't understand what position you are ascribing to me. It might help if you read either the chapter on common sense in http://rebooting.AI or my technical work with Ernie Davis on limits of simulation (links on his web page).
-
-
The 3-body problem in physics is a long-standing problem that reveals the limitations of symbolic methods. There exist no symbolic methods to solve this, this is fundamental.
-
Davis might be arguing for alternative methods to numerical simulation: https://cs.nyu.edu/davise/research.html#Matter … . But there is no evidence in reality for the viability of this approach.
- 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.