Reduction of the dynamic state-space in fuzzy Q-learning.
Szilveszter KovácsPéter BaranyiPublished in: FUZZ-IEEE (2004)
Keyphrases
- state space
- reinforcement learning
- optimal policy
- dynamical systems
- reinforcement learning algorithms
- state variables
- fuzzy logic
- markov chain
- markov decision processes
- heuristic search
- fuzzy sets
- dynamic environments
- multi agent
- function approximation
- particle filter
- dynamic programming
- cooperative
- planning problems
- fuzzy numbers
- partially observable
- fuzzy rules
- fuzzy systems
- search space
- initial state
- learning agent
- continuous state spaces
- state abstraction