Integrating Classical Control into Reinforcement Learning Policy.
Ye HuangChaochen GuXinping GuanPublished in: Neural Process. Lett. (2021)
Keyphrases
- reinforcement learning
- control policy
- optimal policy
- action selection
- control problems
- control policies
- optimal control
- policy search
- control system
- function approximators
- markov decision process
- markov decision processes
- state space
- robotic control
- learning algorithm
- robot control
- partially observable markov decision processes
- real time
- state and action spaces
- continuous state
- partially observable environments
- machine learning
- control strategies
- adaptive control
- function approximation
- control strategy
- dynamic programming
- state dependent
- action space
- partially observable
- markov decision problems
- approximate dynamic programming
- control method
- stochastic control
- agent receives