Kinda funny that @GaryMarcus has a problem with the title of "Solving Rubik's Cube with a Robot Hand" for a result that literally involves a robot hand solving a Rubik's cube.
I will say again that the work itself is impressive, but mischaracterized, and that a better title would have been "manipulating a Rubik's cube using reinforcement learning" or "progress in manipulation with dextrous robotic hands" or similar lines.
Oct 21, 2019 · 2:26 PM UTC
22
11
10
189









