Some Monotonicity Results for Partially Observed Markov Decision Processes.
William S. LovejoyPublished in: Oper. Res. (1987)
Keyphrases
- partially observed
- markov decision processes
- state space
- optimal policy
- finite state
- reinforcement learning
- dynamic programming
- policy iteration
- expected reward
- reinforcement learning algorithms
- partially observable
- transition matrices
- finite horizon
- action space
- planning under uncertainty
- risk sensitive
- decision theoretic planning
- infinite horizon
- model based reinforcement learning
- reachability analysis
- average cost
- decision processes
- reward function
- state abstraction
- state and action spaces
- semi markov decision processes
- factored mdps
- action sets
- decision problems