Title:	Information Processing for Sensor Networks using Distributed Kernel Methods
Date:	October 8, 2008 (Wednesday)
Time:	10:00 a.m. - 11:00 a.m.
Venue:	Room 121, 1/F, Ho Sin-hang Engineering Building, The Chinese University of Hong Kong, Shatin, N.T.
Speaker:	Dr. Anthony Kuh Professor and Chairman Department of Electrical Engineering The University of Hawaii USA

ABSTRACT:

This talk first gives an introduction to kernel methods that have gained popularity in recent years due to a sound theoretical basis, successful applications, and software tools. Kernel methods are based on principles of learning theory and optimization. We give an overview of kernel methods with an emphasis on developing distributed algorithms and subspace methods using quadratic cost functions. By using quadratic cost functions we can exploit many of the results from linear adaptive filter, optimization, and machine learning. We then examine problems of extracting information from sensor networks. We focus on the sensor localization problem, but these methods are also suitable for estimation, detection, and tracking problems involving sensor networks. Here we consider sensor networks where sensors have power and communication constraints. Ad hoc wireless networks are considered where communications between sensors are local. Because of communication constraints we examine a new class of learning algorithms based on distributed processing. The distributed algorithms involve local processing and message passing between different sensors. We demonstrate this algorithm on a sensor network localization problem showing that a distributed kernel learning algorithm can give similar performance to centralized kernel learning algorithms with reduced communication costs. Distributed learning algorithms are closely rated to ensemble learning methods.

BIOGRAPHY:

Anthony Kuh received his B.S. in Electrical Engineering and Computer Science at the University of California, Berkeley in 1979, an M.S. in Electrical Engineering from Stanford University in 1980, and a Ph.D. in Electrical Engineering from Princeton University in 1987. Dr. Kuh previously worked at AT&T Bell Laboratories and has been on the faculty in Electrical Engineering at the University of Hawaii since 1986. He is currently a Professor and Chair of the department. Dr. Kuh's research is in the area of neural networks and machine learning, adaptive signal processing, sensor networks, and communication networks.

Dr. Kuh won a National Science Foundation Presidential Young Investigator Award and is an IEEE Fellow. He was also a recipient of the Boeing A. D. Welliver Fellowship and received a Distinguished Fulbright Scholar's Award working at Imperial College in London. Dr. Kuh was an Associate Editor for the IEEE Transactions on Circuits and Systems, served on the IEEE Neural Networks Administrative Committee, served on the IEEE Neural Networks for Signal Processing Committee, and was a Distinguished Lecturer for the IEEE Circuits and Systems Society. Dr. Kuh co-chaired the 1993 International Symposium on Nonlinear Theory and Its Applications (NOLTA) and served as the technical co-chair for the 2007 IEEE ICASSP both held in Honolulu.

Enquiries: Miss Temmy So at tel 2609 8444

For more information, please refer to http://www.cse.cuhk.edu.hk/seminar