|
Seminar Jointly Organized by NTU, the IEEE Singapore Signal Processing Chapter and PREMIA |
|
|
|
|
Wednesday, 18 June 2008 |
We are pleased to invite you to the following seminar: Kernel-based Class Separability: Theory and Applications by Dr Lei Wang, RSISE, Australian National University. The seminar is jointly organized by the Division of Information Engineering, School of EEE, NTU, the IEEE Singapore Signal Processing Chapter and the Pattern Recognition and Machine Intelligence Association (PREMIA).
TOPIC: Kernel-based Class Separability: Theory and Applications
SPEAKER: Dr Lei Wang, RSISE, Australian National University
DATE: 20 June, Friday 2008
TIME: 11am
VENUE: Executive Seminar Room, S2.2-B2-53, School of EEE, NTU
ABSTRACT
Class separability is a classical concept in the field of pattern recognition. The scatter-matrix based class separability measure has been widely used in discriminant analysis, feature selection, and clustering. This measure can conveniently incorporate the kernel idea introduced in the past few years, giving rising to a kernel-based separability measure. This talk discusses the relationships between this separability measure and the radius-margin bound of the SVMs, the Kernel Alignment criterion, and the Kernel Fisher Discrminant Analysis, providing more insight into the kernel-based class separability measure. Moreover, the above relationships indicate that this separability measure can have a wide range of applications. Its applications to kernel parameter tuning and feature selection are discussed, and its efficiency in handling linearly nonseparable data is demonstrated. This talk is a summarization of our recent research work on the kernel-based class separability in the past few years.
BIOGRAPHY
Dr. Lei Wang received the B.Eng and M.Eng from Southeast University, China in 1996 and 1999, respectively, and the Ph.D. from School of EEE in Nanyang Technological University, Singapore in 2004. He worked as a research associate and research fellow in Nanyang Technological University from 2003 to 2005. After that, he joined the Department of Information Engineering, RSISE, The Australian National University as research fellow. In 2007, he was awarded the Australian Postdoctoral Fellowship by the Australian Research Council. His research interests include computer vision, information retrieval, and machine learning.
This seminar does not require registration. Light Refreshment will be provided. |
