A reinforcement learning with switching controllers for a continuous action space.
Masato NagayoshiHajime MuraoHisashi TamakiPublished in: Artif. Life Robotics (2010)
Keyphrases
- action space
- reinforcement learning
- state space
- control policies
- markov decision processes
- state and action spaces
- continuous state
- continuous state spaces
- function approximation
- reinforcement learning methods
- action selection
- continuous action
- real valued
- control problems
- function approximators
- state action
- machine learning
- reinforcement learning algorithms
- optimal policy
- model free
- optimal control
- markov decision problems
- supervised learning
- dynamic programming
- multi agent
- learning algorithm
- stochastic processes
- single agent
- control system
- temporal difference
- policy search
- markov chain