Regularizing Action Policies for Smooth Control with Reinforcement Learning.
Siddharth MysoreBassel MabsoutRenato MancusoKate SaenkoPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- control policies
- action selection
- optimal policy
- control policy
- control problems
- action space
- fitted q iteration
- control system
- robot control
- optimal control
- markov decision process
- adaptive control
- state action
- state space
- partially observable domains
- learning process
- control method
- neural network
- reward function
- reinforcement learning methods
- control strategy
- decentralized control
- policy search
- robotic control
- control strategies