Structured Threshold Policies for Dynamic Sensor Scheduling - A Partially Observed Markov Decision Process Approach.
Vikram KrishnamurthyDejan V. DjoninPublished in: IEEE Trans. Signal Process. (2007)
Keyphrases
- markov decision process
- partially observed
- optimal policy
- state space
- reinforcement learning
- markov decision processes
- finite horizon
- transition matrices
- infinite horizon
- dynamic environments
- initial state
- transition probabilities
- policy iteration
- reward function
- markov games
- long run
- probabilistic model
- dynamic programming
- long run average cost
- hierarchical reinforcement learning
- stationary policies
- objective function