Control Task for Reinforcement Learning with Known Optimal Solution for Discrete and Continuous Actions.
Michael C. RöttgerAndreas W. LiehrPublished in: J. Intell. Learn. Syst. Appl. (2009)
Keyphrases
- continuous action
- reinforcement learning
- action space
- optimal solution
- policy search
- continuous state
- action selection
- continuous state spaces
- control problems
- continuous state and action spaces
- control policies
- continuous domains
- state space
- discrete space
- optimal control
- np hard
- continuous variables
- control system
- perceptual aliasing
- control policy
- continuous space
- robot control
- robotic control
- discrete data
- function approximation
- continuous data
- discrete variables
- behavioural cloning
- reinforcement learning algorithms
- human operators
- decision theoretic
- control method
- markov decision processes
- control strategies
- state action
- partially observable
- partial observability
- objective function
- linear program
- learning process
- optimal policy
- machine learning
- learning algorithm
- multiagent reinforcement learning
- continuous valued
- knapsack problem
- control strategy
- reinforcement learning methods