Time-in-action RL

Time-in-action RL

The authors propose a novel reinforcement learning (RL) framework, where agent behaviour is governed by traditional control theory. This integrated approach, called time-in-action RL, enables RL to be applicable to many real-world systems, where underlying dynamics are known in their control theoret...

Full description

Bibliographic Details
Main Authors: Jiangcheng Zhu, Zhepei Wang, Douglas Mcilwraith, Chao Wu, Chao Xu, Yike Guo
Format: Article
Language:English
Published: Wiley 2019-02-01
Series:IET Cyber-systems and Robotics
Subjects:
learning (artificial intelligence)
reinforcement learning framework
control theoretical formalism
explicit time function
action value
action time
rl value network
embedded time model
time-in-action rl
Online Access:https://digital-library.theiet.org/content/journals/10.1049/iet-csr.2018.0001

Internet

https://digital-library.theiet.org/content/journals/10.1049/iet-csr.2018.0001

Similar Items