1 paper
Zeyuan Liu, Ziyu Huan, Xiyao Wang +5
Reinforcement learning struggles in the face of long-horizon tasks and sparse goals due to the difficulty in manual reward specification. While existing methods address this by add…