An efficient L2-norm regularized least-squares temporal difference learning algorithm.
Shenglei ChenGeng ChenRuijun GuPublished in: Knowl. Based Syst. (2013)
Keyphrases
- regularized least squares
- temporal difference
- reinforcement learning
- learning algorithm
- reinforcement learning algorithms
- td learning
- function approximation
- evaluation function
- sparse representation
- supervised learning
- monte carlo
- step size
- model free
- machine learning
- learning tasks
- reproducing kernel hilbert space
- action selection
- learning process
- regularization term
- active learning
- learning problems
- training data
- data mining
- graph cuts
- decision trees
- objective function
- data sets
- machine learning algorithms