Search papers, labs, and topics across Lattice.
1
0
2
5
This work proposes residual reinforcement learning as an implicit safety-filtering mechanism for safe humanoid control and shows that the residual policy can act as an implicit safety filter.
