Delayed insertion and rule effect moderation of domain knowledge for reinforcement learning.
Teck-Hou TengAh-Hwee TanPublished in: ADPRL (2013)
Keyphrases
- domain knowledge
- reinforcement learning
- function approximation
- state space
- machine learning
- markov decision processes
- background knowledge
- control knowledge
- reinforcement learning algorithms
- optimal control
- learning problems
- neural network
- optimal policy
- dynamic programming
- association rules
- domain ontology
- prior knowledge
- temporal difference
- multi agent
- temporal difference learning