Truncating Temporal Differences: On the Efficient Implementation of TD(lambda) for Reinforcement Learning
Pawel CichoszPublished in: CoRR (1995)
Keyphrases
- temporal difference
- efficient implementation
- reinforcement learning
- td learning
- function approximation
- reinforcement learning algorithms
- evaluation function
- model free
- action selection
- temporal difference learning
- policy evaluation
- function approximators
- policy iteration
- monte carlo
- step size
- supervised learning
- active set
- state space
- eligibility traces
- machine learning
- fixed point
- multiresolution
- td methods
- temporal difference methods
- markov decision processes
- reinforcement learning methods
- knn
- active learning
- data mining