A Meta Reinforcement Learning Approach for Predictive Autoscaling in the Cloud.
Siqiao XueChao QuXiaoming ShiCong LiaoShiyi ZhuXiaoyu TanLintao MaShiyu WangShijun WangYun HuLei LeiYangfei ZhengJianguo LiJames ZhangPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- function approximation
- cloud computing
- meta level
- learning algorithm
- state space
- model free
- predictive model
- markov decision processes
- direct policy search
- function approximators
- reinforcement learning algorithms
- temporal difference
- multi agent
- machine learning
- learning tasks
- transfer learning
- monte carlo
- supervised learning
- evolutionary algorithm
- action space
- databases
- reinforcement learning methods
- transition model
- policy search
- meta reasoning
- robotic control
- data sets