Search papers, labs, and topics across Lattice.
3
0
6
Achieve safer robot grasping by cleverly crafting adversarial patches that account for both RGB and depth data, outperforming methods that only consider RGB.
Hybrid TD3 stabilizes reinforcement learning in hybrid action spaces by taming overestimation bias with a theoretically grounded weighted clipped Q-learning target.
Robots can now learn manipulation skills from unstructured videos with significantly improved accuracy and generalization by decoupling video understanding from policy learning.
