Authorial Idioms for Target Distributions in TTD-MDPs.
David L. RobertsSooraj BhatKenneth St. ClairCharles Lee Isbell Jr.Published in: AAAI (2007)
Keyphrases
- markov decision processes
- state space
- finite state
- factored mdps
- policy iteration
- optimal policy
- reinforcement learning
- average reward
- model based reinforcement learning
- planning under uncertainty
- finite horizon
- dynamic programming
- infinite horizon
- markov decision process
- probability distribution
- action space
- power law
- decision theoretic planning
- decision diagrams
- random variables
- gaussian distribution
- target tracking
- partially observable markov decision processes
- reward function
- target detection
- heavy tailed
- probabilistic planning
- state and action spaces
- data sets