Dian Wang
@Dian_Wang_
Postdoc @Stanford. Robot learning / geometric deep learning.
478 Following 921 Followers
Long horizon bimanual mobile manipulation requires reasoning in many coordinate frames: base, L/R hands, etc. In which frame would policy work best?
It really depends, so don’t pick. Mixture of Frames Policy: denoise in multiple frame in parallel.
🌐 (1/9)
Show more
