Agreed. Should have said "trained a pair of neural networks to manipulate a Rubik's cube" rather than "solved". That human's also often "solve" Rubik's cubes by implementing a memorised algorithm is no excuse, just argument for Rubik's cube being a crap puzzle.
sorry, but no: you can’t say that the Rubik’s task is “not supposed to be learnable” when the blog opens “we’ve trained a pair of neural networks to solve...” & please read https://twitter.com/garymarcus/status/1185680538335469568?s=21 …, since you missed that the critique is a critique of hype, not the work itself.https://twitter.com/MirowskiPiotr/status/1185860495745699841 …
-
-
- 1 more reply
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.