1 paper · 1 filter
Yongkai Tian, Yirong Qi, Xin Yu +2
In robotic systems, the performance of reinforcement learning depends on the rationality of predefined reward functions. However, manually designed reward functions often lead to p…