Neural Policy Gradient Methods: Global Optimality and Rates of Convergence.

Lingxiao Wang, Qi Cai, Zhuoran Yang, Zhaoran Wang

Browse the full ICLR paper archive.