HONG KONG BAPTIST UNIVERSITY
FACULTY OF SCIENCE
Department of Computer Science Seminar
Learning Compact Features from Multidimensional Data via Tensors
Dr. Haiping Lu
Institute for Infocomm Research
Date: April 8, 2013 (Monday)
Time: 11:30 am - 12:30 pm
Venue: SCT714, Cha Chi Ming Science Tower, Ho Sin Hang Campus
With the advances in sensor, storage and networking technologies, bigger and bigger data are being generated on a daily basis in a wide range of applications. To succeed in this era of big data, it is important to learn compact features for efficient processing. Most big data are multidimensional and can be represented as tensors (multidimensional arrays). This talk focuses on tensor-based learning of compact features from big data. In particular, I will present multilinear subspace learning, a dimension reduction technique developed for tensors.
Multilinear subspace learning directly maps input tensors to a low-dimensional subspace, without reshaping into high-dimensional vectors. It preserves data structure, obtains more compact features, and processes big data more efficiently. I will explain these characteristics using face images and gait sequences. Then, I will study its applications in computer vision, audio signal processing, data mining, neurotechnology, and bioinformatics. Finally, I will discuss my research agenda in learning compact features in an era of big data and small device.
Haiping Lu is a Scientist at the Institute for Infocomm Research, Singapore. He received his Ph.D. degree from the University of Toronto, Canada, in 2008, and the B.Eng. and M.Eng. degrees from Nanyang Technological University, Singapore, in 2001 and 2004, respectively. His research focuses on machine learning and pattern recognition for biometrics, biomedical engineering and mobile computing. He has published 13 journal papers, including 6 IEEE Transactions paper (TNN, TBME, TIFS), and he is a co-inventor of a U.S. patent. His MPCA paper has received more than 160 citations since 2008 (by Google Scholar). He has been a reviewer for 25 international journals, including 10 IEEE Transactions. He is the leading author of the book “Multilinear Subspace Learning: Dimensionality Reduction of Multidimensional Data” (New York: CRC Press, 2013), and his work on EEG has been covered by the ACM TechNews. Dr. Lu received the 2013 IEEE Computational Intelligence Society Outstanding PhD Dissertation Award.
********* ALL INTERESTED ARE WELCOME ***********
(For enquiry, please contact Computer Science Department at 3411 2385)
Department of Computer Science, Hong Kong Baptist University