Safe Exploration of State and Action Spaces in Reinforcement Learning.
Javier GarcíaFernando FernándezPublished in: J. Artif. Intell. Res. (2012)
Keyphrases
- state and action spaces
- reinforcement learning
- markov decision processes
- action space
- state space
- markov decision problems
- action selection
- partially observable markov decision process
- real valued
- average reward
- optimal policy
- reinforcement learning algorithms
- function approximation
- partially observable
- partially observable markov decision processes
- neural network
- multi agent
- average cost
- steady state
- multi agent systems
- optimal solution