Toward Designing Cost-Optimal Policies to Utilize IaaS Clouds with Online Learning.
Xiaohu WuPatrick LoiseauEsa HyytiäPublished in: IEEE Trans. Parallel Distributed Syst. (2020)
Keyphrases
- optimal policy
- online learning
- average cost
- cloud computing
- markov decision processes
- lost sales
- decision problems
- reinforcement learning
- long run
- finite state
- finite horizon
- state space
- infinite horizon
- average reward
- dynamic programming
- multistage
- dynamic programming algorithms
- expected cost
- e learning
- policy iteration
- total cost
- serial inventory systems
- state dependent
- average reward reinforcement learning
- control policies
- asymptotically optimal
- sample path
- fixed cost
- active learning
- finite number
- initial state
- sufficient conditions
- inventory models
- lot size
- long run average cost