Acceleration of Reinforcement Learning with Incomplete Prior Information.
Kento TerashimaHirotaka TakanoJunichi MurataPublished in: J. Adv. Comput. Intell. Intell. Informatics (2013)
Keyphrases
- prior information
- reinforcement learning
- prior knowledge
- bayesian inference
- state space
- model free
- hyperparameters
- function approximation
- reinforcement learning algorithms
- prior models
- uncertain information
- markov decision processes
- statistical information
- learning process
- missing values
- dynamic programming
- autonomous learning
- data sets
- prior distribution
- prior probabilities
- support vector
- neural network