Edge-Assisted Secure and Dependable Optimal Policies for the 5G Cloudified Infrastructure.
Glaucio H. S. CarvalhoIsaac WoungangAlagan AnpalaganIssa TraoréPeriklis ChatzimisiosPublished in: ICC (2022)
Keyphrases
- optimal policy
- markov decision processes
- decision problems
- state space
- finite horizon
- reinforcement learning
- dynamic programming
- infinite horizon
- multistage
- sufficient conditions
- finite state
- state dependent
- policy iteration
- long run
- dynamic programming algorithms
- average reward
- average reward reinforcement learning
- serial inventory systems
- bayesian reinforcement learning
- initial state
- markov decision process
- average cost
- control policies
- lost sales
- inventory level
- markov decision problems
- continuous monitoring
- dynamical systems
- search space