On-line discovery of temporal-difference networks.
Takaki MakinoToshihisa TakagiPublished in: ICML (2008)
Keyphrases
- temporal difference
- reinforcement learning
- td learning
- function approximation
- evaluation function
- monte carlo
- temporal difference learning
- model free
- reinforcement learning algorithms
- step size
- action selection
- data mining
- function approximators
- genetic algorithm
- least squares
- machine learning
- policy iteration
- active learning
- multiscale
- feature extraction
- data sets
- image compression
- state space
- multi objective