报告题目: Bayesian Ying-Yang System, Best Harmony Learning, and Five Action Circling
报告人：Lei Xu
时间(Time): 4:00pm, May 28, 2013, Tuesday
地点 (Venue): 信息学院一楼会议室
报告人简介：Lei Xu, chair professor of Chinese Univ Hong Kong (CUHK), Fellow of IEEE (2001-), IAPR Fellow (2002-), and Academician of European Academy of Sciences (2002-).  He completed his Ph.D thesis at Tsinghua Univ by the end of 1986, became postdoc at Peking Univ in 1987, then promoted to associate professor in 1988 and a professor in 1992. During 1989-93 he was research associate and postdoc in Finland, Canada and USA, including Harvard and MIT. He joined CUHK as senior lecturer in 1993, professor in 1996, and chair professor in 2002. He has published about 100 journal papers, with a number of well-cited papers on neural networks, statistical learning, and pattern recognition, e.g., his papers got over 4000 citations according to SCI and over 9000 citations according to Google Scholar (GS), with the citations of top-10 papers being (SCI: 911, 344, 285, 204, 184, 174, 144, 96, 96, 85) and (GS: 1949, 744, 587, 511, 467, 299, 271, 209, 202, 191). Prof. Xu served a governor of international neural network society (INNS), a past president of APNNA, and a member of Fellow committee of IEEE Computational Intelligence Society. He is the founding editor-in-chief of Springer open access Journal Applied Informatics, and also serves or served as an associate editor for eight academic journals. Prof. Xu has received several national and international academic awards (e.g., 1993 National Nature Science Award, 1995 INNS Leadership Award and 2006 APNNA Outstanding Achievement Award). 

报告内容(About the Talk:): Proposed in 1995 and systematically developed over fifteen years, Bayesian Ying-Yang (BYY) learning is a statistical approach for an intelligent system via two complementary Bayesian representations of a joint distribution on the external observation X and its inner representation R, called BYY system. A Ying-Yang best harmony principle is proposed for learning all the unknowns in the system, with help of an implementation featured by a five action circling. BYY learning provides not only a general framework that accommodates typical learning approaches from a unified perspective but also a road that leads to improved model selection criteria and automatic model selection embedded learning. This talk introduces fundamentals of BYY learning and typical learning algorithms, in a comparison with other algorithms, and illustrated via applications on clustering analyses, image segmentation, speech recognition, HRRP-based target recognition, and tasks of gene analyses.