Spatially Controlled Relay Beamforming: $2$-Stage Optimal Policies.
Dionysios S. KalogeriasAthina P. PetropuluPublished in: CoRR (2017)
Keyphrases
- optimal policy
- markov decision processes
- decision problems
- finite state
- long run
- state space
- finite horizon
- reinforcement learning
- dynamic programming
- infinite horizon
- multistage
- average reward
- state dependent
- sufficient conditions
- dynamic programming algorithms
- policy iteration
- bayesian reinforcement learning
- serial inventory systems
- markov decision process
- average cost
- single stage
- initial state
- control policies
- inventory control
- markov decision problems
- semi markov decision processes
- partially observable markov decision processes
- expected reward
- average reward reinforcement learning
- reward function