Enhancing reinforcement learning by a finite reward response filter with a case study in intelligent structural control.
Hamid Radmard RahmaniCarsten KönkeMarco A. WieringPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- control problems
- optimal control
- control policy
- function approximation
- state space
- adaptive control
- case study
- control system
- optimal policy
- control strategies
- robotic control
- average reward
- learning capabilities
- structural information
- control method
- finite number
- reinforcement learning methods
- robot control
- reinforcement learning algorithms
- markov decision processes
- test bed
- intelligent systems
- state and action spaces
- markov chain
- control strategy
- temporal difference
- decision support
- state action
- model free
- policy gradient
- long run
- total reward
- image processing