Symbolic Magnifying Lens Abstraction in Markov Decision Processes.
Pritam RoyDavid ParkerGethin NormanLuca de AlfaroPublished in: QEST (2008)
Keyphrases
- markov decision processes
- decision theoretic planning
- state abstraction
- finite state
- state space
- optimal policy
- reinforcement learning
- dynamic programming
- reachability analysis
- partially observable
- finite horizon
- transition matrices
- policy iteration
- decision processes
- average cost
- macro actions
- infinite horizon
- markov decision process
- reinforcement learning algorithms
- action space
- planning under uncertainty
- probabilistic planning
- factored mdps
- state and action spaces
- semi markov decision processes
- model based reinforcement learning
- decision diagrams
- risk sensitive
- long run
- function approximation
- markov chain