Search papers, labs, and topics across Lattice.
Affiliation:
1
0
2
LG-AC outperforms traditional goal-conditioned methods by explicitly leveraging intermediate goals, tackling the notorious challenges of goal chaining in reinforcement learning.
