Autonomous nondeterministic tour guides: improving quality of experience with TTD-MDPs.
Andrew S. CantinoDavid L. RobertsCharles L. Isbell Jr.Published in: AAMAS (2007)
Keyphrases
- markov decision processes
- finite state
- decentralized control
- state space
- reinforcement learning
- policy iteration
- optimal policy
- decision theoretic planning
- dynamic programming
- partially observable
- markov decision process
- high quality
- fully observable
- higher quality
- planning under uncertainty
- average reward
- finite horizon
- quality assessment
- quality measures
- factored mdps
- learning algorithm
- decision diagrams
- infinite horizon
- state and action spaces
- autonomous navigation
- average cost
- reward function
- neural network
- robotic systems
- cooperative
- machine learning