1 paper
Banafsheh Rafiee, Sina Ghiassian, Jun Jin +3
In this paper, we explore an approach to auxiliary task discovery in reinforcement learning based on ideas from representation learning. Auxiliary tasks tend to improve data effici…