Flink ML is a stream batch machine learning algorithm library based on Apache Flink, and is a subproject of Apache Flink. In this report, we will introduce the new features of Flink ML 2.2.0 and the landing scene of Flink ML in Alibaba.

• Infrastructure to support online inference services
• Rich feature engineering algorithms
• Design of online learning algorithm and its application in real-time log clustering scenario

Speakers:

Zhang Zhipeng: Aliyun Intelligence, Algorithm expert, Alibaba algorithm expert, Apache Flink committer

Dr. Zhipeng Zhang graduated from Peking University with a research focus on distributed machine learning. After graduation, he joined Alibaba Cloud machine learning team PAI, mainly engaged in the design, development and improvement of Flink ML.


Hongfan: Alibaba, Algorithm expert, Dr. Hong Fan graduated from Peking University. After graduation, I joined the machine learning team of Alibaba, mainly engaged in the development and improvement of machine learning related to Flink, and was a Flink ML Contributor. He was also involved in the development of Alink, the previous generation of Flink machine learning library.