Erlang planning network: An iterative model-based reinforcement learning with multi-perspective

作者：

Highlights：

• Multi-perspective fusion attenuates the impact of model error in model-based reinforcement learning.

• Bi-level architecture is used to integrate decisions under multiple views.

• Good generality for different kinds of robot control tasks.

• Dramatically improve performance without more sampling in the environment.

摘要

•Multi-perspective fusion attenuates the impact of model error in model-based reinforcement learning.•Bi-level architecture is used to integrate decisions under multiple views.•Good generality for different kinds of robot control tasks.•Dramatically improve performance without more sampling in the environment.

论文关键词：Model-based reinforcement learning,Multi-perspective,Bi-level,Planning,Trajectory imagination

论文评审过程：Received 31 July 2021, Revised 1 December 2021, Accepted 23 March 2022, Available online 25 March 2022, Version of Record 31 March 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108668