Accelerated Q-learning for fail state and action spaces.
In-Won ParkJong-Hwan KimKui-Hong ParkPublished in: SMC (2008)
Keyphrases
- state and action spaces
- reinforcement learning
- state space
- markov decision processes
- action space
- reinforcement learning algorithms
- optimal policy
- markov decision problems
- policy iteration
- function approximation
- average reward
- partially observable markov decision process
- action selection
- model free
- dynamic programming
- multi agent
- heuristic search
- learning algorithm
- optimal control
- markov chain
- decision making
- planning problems
- partially observable
- state variables
- decision processes
- neural network
- dynamical systems