Abstraction from demonstration for efficient reinforcement learning in high-dimensional domains.
Luis C. CoboKaushik SubramanianCharles Lee Isbell Jr.Aaron D. LantermanAndrea Lockerd ThomazPublished in: Artif. Intell. (2014)
Keyphrases
- high dimensional
- reinforcement learning
- multi dimensional
- multi agent
- low dimensional
- state abstraction
- high level
- similarity search
- partially observable domains
- robotic control
- decision theoretic planning
- cost effective
- application domains
- high dimensional data
- computationally efficient
- nearest neighbor
- state space