1 paper · 1 filter
Abhinav Agarwal, Adam Wei, Taylan Kargin +6
Imitation learning has enabled highly-dexterous robotic manipulation from RGB observations. Policies trained with these methods, however, typically condition robot actions on only…