A Nonparametric Off-Policy Policy Gradient

Understand

Reinforcement learning (RL) algorithms still suffer from high sample complexity despite outstanding recent successes.

• The need for intensive interactions with the environment is especially observed in many widely popular policy gradient algorithms that perform updates using on-policy samples.
• The price of such inefficiency becomes evident in real-world scenarios such as interaction-driven robot learning, where the success of RL has been rather limited.
• We address this issue by building on the general sample efficiency of off-policy algorithms.