Reward Shaping with Dynamic Trajectory Aggregation.

Takato Okudo, Seiji Yamada

Browse the full IJCNN paper archive.