Search papers, labs, and topics across Lattice.
1
0
3
Reinforcement learning agents can now learn to be "good" (i.e., norm-compliant) via a novel pipeline that leverages argumentation-based normative advisors and automatically extracts the reasoning behind those norms.
