An adjustment method of the number of states of Q-learning segmenting state space adaptively.
Tomoki HamagamiHironori HirataPublished in: SMC (2003)
Keyphrases
- state space
- high accuracy
- dynamic programming
- reinforcement learning
- small number
- machine learning
- data sets
- high precision
- detection method
- state action
- state variables
- support vector machine
- computational cost
- experimental evaluation
- cost function
- pairwise
- probabilistic model
- k means
- preprocessing
- clustering method
- computational complexity
- similarity measure
- decision trees
- genetic algorithm