Exploring Continuous Action Spaces with Diffusion Trees for Reinforcement Learning.
Christian VollmerErik SchaffernichtHorst-Michael GrossPublished in: ICANN (2) (2010)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- continuous state
- real valued
- continuous state spaces
- stochastic processes
- reinforcement learning methods
- continuous action
- control policies
- state action
- single agent
- action selection
- model free
- function approximators
- heuristic search
- reinforcement learning algorithms
- function approximation
- markov chain
- skill learning
- partially observable
- control problems
- markov decision process
- random variables
- optimal policy
- dynamic environments
- dynamic programming
- machine learning