Sorry to break it to you but this is factually incorrect. It didn't learn to solve the cube at all. It was told which moved to make by special software - what it learned to do was to manipulate the cube to make the moves. Impressive but very different. @GaryMarcushttps://twitter.com/washingtonpost/status/1185384618196946944 …
-
Wait wasn’t there also a human solver present?
no human, just an old symbolic algorithm. hybrids exist (deep learning + monte carlo tree search) but they didn’t use them
-
-
Symbolic algorithm + deep reinforcement learning is as hybrid as you can get. It's about as general as a Tic-tac-toe program. You should be happy, Marcus.
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.