1 paper · 1 filter
Jin Li, Yue Wu, Mengsha Huang +3
Recurrent neural policies are widely used in partially observable control and meta-RL tasks. Their abilities to maintain internal memory and adapt quickly to unseen scenarios have…