LC-Learning: Phased Method for Average Reward Reinforcement Learning - Analysis of Optimal Criteria.
Taro KondaTomohiro YamaguchiPublished in: PRICAI (2002)
Keyphrases
- average reward reinforcement learning
- prior knowledge
- synthetic data
- high accuracy
- learning mechanism
- dynamic programming
- cost function
- similarity measure
- high precision
- experimental evaluation
- significant improvement
- preprocessing
- active learning
- learning scheme
- exhaustive search
- learning tasks
- reinforcement learning
- detection method
- theoretical analysis
- data analysis
- unsupervised learning
- computational complexity
- classification method
- worst case
- probabilistic model
- computational cost
- learning process
- training data
- weighting coefficients
- machine learning