报告题目：LARGE-SCALE TIME-SERIES CLUSTERING WITH k-ARs

报 告 人：岳作功 博士后（卢森堡大学）

报告时间：2019年10月24日 上午10:30—11:30

报告地点：南一楼中314

报告摘要：Time-series clustering involves grouping homogeneous time series together based on certain similarity measures. The mix- ture AR model (MxAR) has already been developed for time series clustering, as has an associated EM algorithm. How- ever, this EM clustering algorithm fails to perform satisfacto- rily in large-scale applications due to its high computational complexity. This paper proposes a new algorithm, k-ARs, which is a limiting version of the existing EM algorithm. It shows remarkably good computational performance when ap- plied to large-scale clustering problems as illustrated on some benchmark simulations motivated by some real applications.

报告人简介：Zuogong YUE (岳作功) works as postdoctoral fellow in the School of Electrical Engineering and Telecommunications, the University of New South Wales (UNSW), Sydney, Australia. He received a bachelor degree in mechatronics engineering from Zhejiang University, and a master degree in mechanical engineering from Hong Kong University of Science and Technology. He joined the Group of Systems Control in Luxembourg Centre for Systems Biomedicine in 2014 and received his engineering degree in 2018. His research interests focus on system identification, network reconstruction and signal processing, with techniques adopted/modified from convex optimization, statistics, stochasticanalysis and machine learning, including but not limited to, Bayesian analysis, sampling methods, multivariate statistics and etc.