Testing a Q-learning approach for derivation of scaling policies in cloud-based applications.
Yue JinMakram BouzidDimitre KostadinovArmen AghasaryanPublished in: ICIN (2018)
Keyphrases
- optimal policy
- cooperative
- function approximation
- reinforcement learning
- hierarchical reinforcement learning
- cloud computing
- multi agent
- state space
- model free
- learning algorithm
- discounted reward
- reinforcement learning algorithms
- reward function
- action selection
- test cases
- decision problems
- markov decision processes
- test set
- reinforcement learning methods
- policy search
- dynamic programming
- machine learning