Reinforcement learning of motor skills using Policy Search and human corrective advice.
Carlos CeleminGuilherme MaedaJavier Ruiz-del-SolarJan PetersJens KoberPublished in: Int. J. Robotics Res. (2019)
Keyphrases
- action selection
- reinforcement learning
- policy search
- motor skills
- continuous state
- reinforcement learning algorithms
- function approximation
- state space
- learning algorithm
- machine learning
- markov decision processes
- everyday life
- humanoid robot
- reward function
- optimal control
- markov decision problems
- reinforcement learning methods
- policy gradient
- dynamic programming
- control strategies