Magnifying-Lens Abstraction for Markov Decision Processes.
Luca de AlfaroPritam RoyPublished in: CAV (2007)
Keyphrases
- markov decision processes
- decision theoretic planning
- state abstraction
- state space
- finite state
- reinforcement learning
- optimal policy
- policy iteration
- infinite horizon
- factored mdps
- planning under uncertainty
- reachability analysis
- dynamic programming
- decision processes
- state and action spaces
- average reward
- finite horizon
- partially observable
- macro actions
- risk sensitive
- average cost
- reinforcement learning algorithms
- transition matrices
- action space
- semi markov decision processes
- model based reinforcement learning
- probabilistic planning
- markov decision process
- action sets