电子信息学院学术报告:Subspace Clustering of Data from Single and Multiple Views

SpeakerIvica Kopriva

Title: Subspace Clustering of Data from Single and Multiple Views

Time: 930 AM, Nov. 23th

Location: Room 306, Electronics and Information Building, Tiancizhuang Campus

Abstract

In many applications high-dimensional data points are well represented by a union of low-dimensional linear subspaces. Subspace clustering refers to assignment of data points or patterns to subspaces they are drawn from. That is achieved by k-means clustering of eigenvectors of Laplacian matrix which itself is obtained from representation matrix. As opposed to clustering algorithms that rely on spatial proximity between data points subspace clustering handles clusters of arbitrary shapes. However, two issues are critical in subspace clustering are: (i) good representation matrix is necessary for accurate clustering; (ii) computational complexity of spectral clustering is cubic in terms of number of data points and that prevents its application to clustering of large scale datasets. The talk will present original results of the speaker and his collaborators that address both issues. To address first issue class of algorithms is developed for estimation of the low-rank sparse representation matrix from data. To address the second issue family of nonlinear (kernelized) orthogonal nonnegative matrix factorization algorithms is derived. They are equivalent to nonnegative spectral clustering but have close to quadratic computational complexity.

Speaker Biography:

Ivica Kopriva obtained PhD degree from the Faculty of Electrical Engineering and Computing, University of Zagreb in 1998 with a subject in blind source separation. From 2001 till 2005 he was research and senior research scientist at Department of Electrical and Computer Engineering, The George Washington University, Washington D.C., USA. Since 2006 he is senior scientist at the Ru?er Bo?kovi? Institute, Zagreb, Croatia. His research interests are related to development of algorithms for unsupervised learning with applications in biomedical image analysis, chemometrics and bioinformatics. He published over 40 papers in internationally recognized journals and holds 3 US patents. He is co-author of the research monograph: Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised and Unsupervised Learning, Springer Series: Studies in Computational Intelligence, 2006. He is senior member of the IEEE and the OSA.