Controlling the Continuos Positive Airway Pressure-Device Using Partial Observable Markov Decision Processes.
Clemens KreutzJosef HonerkampPublished in: HPSC (2003)
Keyphrases
- markov decision processes
- finite state
- dynamic programming
- optimal policy
- state space
- transition matrices
- partially observable
- reinforcement learning
- decision theoretic planning
- reinforcement learning algorithms
- policy iteration
- average reward
- planning under uncertainty
- finite horizon
- factored mdps
- model based reinforcement learning
- reachability analysis
- action sets
- infinite horizon
- state and action spaces
- risk sensitive
- markov decision process
- machine learning
- long run
- decision processes
- semi markov decision processes
- average cost