XU, Lei |
|
Lei Xu is a
professor of Computer Science and Engineering Chinese Univ Hong Kong (CUHK)
where he first joined as a senior lecturer in 1993, became a professor in 1996
and then took the current chair professor position in 2002. He completed his
Ph.D thesis at Tsinghua Univ by the end of 1986, then joined Dept. Math, Peking
Univ in 1987 first as a postdoc and then was exceptionally promoted to
associate professor in 1988 and to a full professor in 1992. During 1989-93, he
worked at several universities in
Prof. Xu has published dozens of journal papers and also many papers in conference proceedings and edited books, covering the areas of Statistical Learning and Neural Networks, Financial Engineering, Computer Vision and Pattern Recognition, Signal Processing, and Artificial Intelligence. Also, he has given over 50 keynote /invited / tutorial lectures at major international conferences on Computational Intelligence and Neural Networks, and has been internationally known with his several well-cited contributions on adaptive PCA and independence learning, classifier combination and mixture model based learning, rival penalized competition, etc, as well as his Bayesian Ying-Yang unified statistical learning system and theory. Also, he and his colleague E.Oja's invention on Randomized Hough Transform has a widely impact in the field of pattern recognition and computer vision too. E.g., his papers got over 7100 citations according to Google Scholar (GS) and over 3900 citations according to SCI, with the first 10 papers scored over 4600 (GS) and 2310 (SCI). One single paper has scored 1690 (GS) and 857(SCI). Also, the H-index is 34 (GS) and 27 (SCI).
Prof. Xu has served as an associate
editor for eight academic journals, especially on neurocomputing, e.g.,
including Neural Networks (1995-present) and IEEE Trans. on Neural Networks
(1994-98). Moreover, he has served as a governor on the Governing Board of International Neural Networks
Society (INNS) (2001-03), as well as the INNS award committee (2002-03)
and the Fellow committee of IEEE Computational
Intelligence society (06), the chair of Computational Finance Technical
Committee of IEEE Neural Networks Society (2001-03), and a past president of
Asian-Pacific Neural Networks Assembly. He is also served as a general chair of
IEEE Intl Conf. on CIFER03, Intl Confs IDEAL98, IDEAL00, Joint-ICANN-ICONIP03
Program Committee co-chair, a program committee chair of ICONIP96 and a
honorary chair of ICONIP06, and the chairs of two NIPS Workshops, as well as
program/organizing/Advisory committee members on major world conferences on
Neural Networks, including WCNN, ICNN, IJCNN, IEEE WCCI, ICONIP, NIPS, ICANN,
IDEAL, and Intl. Conf. on Computational Finance, etc. Prof. Xu has also been
serving as a member of engineering panel, Hong Kong RGC research committee
(2001-06), a member of selection committee, Chinese NSFC/HK RGC Joint research
scheme (2002-05, 08), external expert for Chinese NSFC information science
panel (2004-06), and an nominator for the prestigious Kyoto prize (2003,2007).
Prof. Xu is a Fellow of IEEE
(2001- ), Fellow of IAPR (International Association for Pattern
Recognition) (2002-), member of European
Academy of Sciences (2002-). He has been listed in many international
biographical publications. Prof Xu has received an 1995 International Neural
Networks Society Leadership Award and several Chinese national prestigious
academic awards, including 1993 Chinese National Nature Science Award,
1988 Chinese State Education Council FOK YING TUNG Award, and the second
of the 10 winners of the 1988 BEIJING YOUNG SCIENTISTS PRIZE. Following
the 2004 winner Prof Amari and the 2005 winner Prof. Fukushima, Prof. Xu has received the 2006 APNNA Outstanding Achievement Award.
Address: Department of Computer Science and
Engineering, The Chinese
Phone: +852-26098423 Email: lxu[at]
cse[dot]cuhk[dot]edu[dot]hk
Research Interests
- STATISTICAL LEARNING (unsupervised, supervised, unified theory,
model selection) and applications to neural computing, financial
engineering & time series, signal separation, bioinformatics).
- COMPUTER VISION (Random Hough Transform, detection of shape and
motion of multiple objects).
Representative
Publications (by topics)
·
Overview Papers in Journals/Encyclopedia/Handbook
(enter)
Bayesian Ying-Yang system and best harmony learning
Randomized Hough
Transform
Rival Penalized
Competitive Learning
Combining Classifiers and Learning Mixture-of-Experts
Independent Subspaces
RBF Functions and Subspace Based Functions
·
Bayesian Ying-Yang System and
Harmony Learning Theory (enter)
·
Unsupervised Learning (enter)
Least MSE Reconstruction, PCA, MCA and Nonlinear Extensions
Competitive Learning and
Clustering analysis
Gaussian Mixture and EM Algorithm
Bayesian Approach
·
Independent component analysis
and independent factor Analysis
(enter)
·
Supervised Learning (enter)
Multiple Experts and Classifier Combination
Radial Basis
Functions and
Subspace Based Functions
Feed-forward
Networks
·
Randomized
Hough Transform and Object Detection (enter)
·
Financial Engineering and
Signal Processing (enter)
·
Combinatorial Optimization (enter)
Representative Publications (by years) (enter)
Three Special issues on Frontiers of Electrical and Electronic Engineering in
China (as Guest Editor)
· Volume 5, Number 3 / September 2010, consisting of nine invited survey articles written by
Shun-ichi Amari, Erika Oja, Jorma Rissanen, Alan Yuille Jürgen Schmidhuber, Raymond W. Yeung, Runsheng Chen, Yanda Li, and Lei Xu, on Emerging themes on information theory and Bayesian Approach, especially on topics the authors pioneered.
All the authors are IEEE Fellows or/and Members of National Academy of Sciences.
See http://www.springerlink.com/content/1673-3460/5/3/
· Volume 6, Number 1 / March 2011, A special issue on Machine learning and intelligence science: Sino-foreign interchange workshop IScIDE2010 (A). consisting of fourteen invited survey or research articles. This special issue is bisected.
o
The first half provides an insightful view for
the core achievements by Chinese colleagues, with articles written by professors
from major universities in mainland
• Describing data samples with
reduced dimensionality;
• Integrating
multi-sources of partial information for a better performance.
o
The second half aims at a much wider scope of
intelligence science (vision system, neural system, human behavior, and generic
issues) for broadening the readers’ knowledge of this journal, written by
professors from Hong Kong,
See http://www.springerlink.com/content/1673-3460/6/1/
·
Volume 6, Number 2 / June 2011, A special issue
on Machine learning and intelligence science: Sino-foreign
interchange workshop IScIDE2010
(B). consisting of nineteen invited research articles: the two-third comes from
major universities in mainland of
See http://www.springerlink.com/content/1673-3460/6/2/