An Improved Trust-Region Method for Off-Policy Deep Reinforcement Learning.

Hepeng Li, Xiangnan Zhong, Haibo He

Browse the full IJCNN paper archive.