Shaping in reinforcement learning by knowledge transferred from human-demonstrations of a simple similar task.
Guo-fang WangZhou FangPing LiPublished in: J. Intell. Fuzzy Syst. (2018)
Keyphrases
- reinforcement learning
- domain knowledge
- human knowledge
- knowledge base
- knowledge acquisition
- knowledge discovery
- state space
- knowledge extraction
- knowledge transfer
- prior knowledge
- neural network
- knowledge sharing
- human experts
- function approximation
- optimal policy
- human cognition
- complex domains
- learned knowledge
- reward shaping
- human users
- learning systems
- knowledge based systems
- case study
- decision making
- machine learning