講題: AI Application on Anomaly Detection/Prediction

時間日期地點: 3/6 (Wed) 3:30-5:00PM in ED 117

講員名字&簡歷:

Phone Lin, IEEE Fellow, Professor, Dept of Computer Science and Information Engineering, National Taiwan University, Taiwan.

Phone Lin (M’02–SM’06–F’17) received the B.S. and Ph.D. degrees in computer science and information engineering from National Chiao Tung University, Hsinchu, Taiwan, in 1996 and 2001, respectively. He is a Professor with National Taiwan University, Taiwan, holding a professorship within the Department of Computer Science and Information Engineering, Graduate Institute of Networking and Multimedia, and Telecommunications Research Center, College of EECS, and Graduate Institute of Medical Device and Imaging of College of Medicine. He is also a Researcher (joint appointment) and the Deputy Director with National Science and Technology Center for Disaster Reduction, Taiwan. He serves on the editorial boards of several journals, such as the IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, the IEEE Network Magazine, the IEEE INTERNET OF THINGS JOURNAL, and Computer Networks Journal. He was the Chair of IEEE Vehicular Technology Society Taipei Chapter from 2014 to 2015. He was a recipient of many prestigious research awards, such as the Outstanding Research Award, Ministry of Science and Technology, Taiwan, in 2016, and the Best Young Researcher of IEEE ComSoc Asia–Pacific Young Researcher Award in 2007. He is an ACM Senior Member..

大綱:

Anomaly detection/prediction usually relies on wide domain knowledge to build up the tools to automatically detect/predict abnormal events or behaviors of an IoT system. An IoT system may consist of the machines with different capabilities, functionalities and ages. Abnormal events or behaviors are usually rare events. It is time-consuming and high-cost to build up the domain knowhow of the IoT systems and collect enough data points of the anomaly. In this talk, I first identify the issues and challenges. Then I illustrate a general environment for anomaly detection/perdition. Then I will illustrate the core technologies and exiting solutions that may be applied for anomaly detection/prediction. I identify what cannot be achieved by the existing solutions.