报告题目：Matrix optimization models and algorithms for Data Clustering
Abstract: Clustering is to classify data into groups according to a predefined distance or similarity measure. It has wide applications in data mining, pattern recognition, image processing and other machine learning areas. It is well known that lots of clustering models, like K-means and K-indicators, can be written as non-convex matrix optimization problems.
In this work, we attempt to employ the classical optimization algorithms to solve the unsupervised clustering task. Numerical examples on several benchmark datasets are conducted to evaluate the efficiency and accuracy of our approach.