报告题目：Reinforcement Learning for Cloud Intelligence

报告人：王露 研究员（微软亚洲研究院MSRA）

主持人：王祥丰 副教授

报告时间：2023年5月8日（星期一）10:00-11:30

报告地点：普陀校区理科楼B112室

报告摘要：

AIOps in the cloud requires safe, interpretable, and efficient decision making, especially for tasks like disk failure mitigation, vCPU oversubscription, and VM allocation. At Microsoft, we're using reinforcement learning to address these challenges on various cloud platforms. To ensure safe decision making, we propose chance constraint RL methods that maximize benefits while satisfying cost constraints. For efficient decision making, we're developing cascade RL, which quickly filters out unnecessary actions. Additionally, we're exploring logic rule learning-based decision making to improve interpretability.

报告人简介：

Lu Wang is a researcher at MSRA specializing in reinforcement learning, cloud intelligence, healthcare, finance, and other related areas. She received her PhD degree in ECNU and has published over 20 papers on RL, including ICML, ICLR, and KDD.