Reinforcement Learning via Fenchel-Rockafellar Duality
Xuhui Liu (NJU)
Short Abstract: In this talk, we review basic concepts of Fenchel-Rockafellar duality. We summarize how this duality may be applied to a variety of RL settings, including policy evaluation or optimization, online or offline learning.
Reference: https://arxiv.org/abs/2001.01866