Mixed Observability Markov Decision Processes for Overall Network Performance Optimization in Wireless Sensor Networks.
Daniel L. KovacsWuyungerile LiNaoki FukutaTakashi WatanabePublished in: AINA (2012)
Keyphrases
- markov decision processes
- wireless sensor networks
- partially observable
- state space
- finite state
- optimal policy
- transition matrices
- policy iteration
- planning under uncertainty
- reinforcement learning
- decision processes
- infinite horizon
- reinforcement learning algorithms
- dynamic programming
- average reward
- model based reinforcement learning
- reachability analysis
- risk sensitive
- finite horizon
- decision theoretic planning
- action space
- factored mdps
- routing protocol
- markov decision process
- sensor networks
- partially observable markov decision processes
- average cost
- action sets
- reward function
- markov chain
- decision diagrams