Content
@
0 reply
0 recast
0 reaction
ππͺπΎπ‘π‘πΎ
@gm8xx8
They trained a diffusion policy using 26,000 demonstrations across 5 tasks on the Aloha 2 robot, relying solely on imitation learning. The model has 217 million parameters, and training involved 2 million steps with 64 TPUv5e. Absolute π₯ π: https://openreview.net/pdf?id=gvdXE7ikHI π: https://x.com/remicadene/status/1832911834208149683?s=46
1 reply
0 recast
7 reactions
bubble
@buq
All things are free poetry @drya
0 reply
0 recast
0 reaction
