报 告 人： 刘一鸣 博士
报告题目：High dimensional clustering Covariance clustering for mixture data
Clustering is an essential subject in unsupervised learning. It is a common technique used in many fields, including machine learning, statistics, bioinformatics, and computer graphics. To classify different samples into a homogeneous group, it is based on different criterions. In this paper, we focus on the clusters that are characterized by the different covariances, and we study the clustering method for the high dimensional mixtures. According to this setting, we propose a new approach,covariance clustering method, to conduct clustering. Both theoretical and numerical properties of the covariance clustering method are discussed.Specifically, we propose one algorithm that is applicable to do the clustering in different settings. In addition, we prove that the misclustering error for this algorithm converges to zero with probability tends to one under mild conditions. Simulation studies also demonstrate that the covariance clustering method outperforms other methods under a variety of settings.