Shoujue Wang
Institute of Semiconductors, Chinese Academy of Sciences, Beijing, China
wsjue@red.semi.ac.cn

Outline

1. An Introduction to High Dimensional Space (HDS) Point-set Covering Recognition (PCR or named Biomimetic Pattern Recognition)
2. Computational Descriptive Geometry in High Dimensional Space as a Tool in Point-set Covering Recognition (PCR)
3. A Short Description on Multi-weighted Neural Networks Used in PCR
4. Comparison Results of Recognition Rates Between PCR and SVM as well as HMM in some Application Fields

Speaker
Prof. Wang Shoujue, specialist on information science, was born in June 1925 in Shanghai. He graduated from the Department of Electrical Engineering, Tong Ji University in 1949. After graduation, he served as Research Assistant in the Institute of Radium, Peking Academy in Shanghai. He joined the Institute of Applied Physics, Chinese Academy of Sciences in Beijing in 1956, and served as director of the Semiconductor Devices Laboratory since the founding of the Semiconductor Institute, CAS in 1960. He became deputy director and director of the Semiconductor Institute, CAS in 1977 and 1983 respectively. He was elected the Member of Chinese Academy of Sciences in 1980. He is now a research fellow in the Semiconductor Institute, CAS, in charge of the research laboratory on the Artificial Neural Networks and Machine Thinking in Image. He is also a guest Professor at the Tong Ji University in Shanghai and the Zhejiang University of Technology in Hangzhou.
Professor Wang developed the first domestic made high frequency switching transistor in 1958, for using in the early high speed transistorized computer, for computing in research works on the nuclear physics. He also developed the earliest silicon planar transistors and solid state circuits in China in 1963 and 1965, the Pattern Generator for LSI mask making in 1971. By using his invention on the Integrated High Speed Fuzzy Logic Circuit DYL, published in 1978, he reduced the converting time of 8 bit D/A converter chips form 80 ns to 4 ns. In the last decade of the twentieth century, he worked and got significant achievements on Artificial Neural Networks, including chips, hardware, mathematical models, algorithms and applications in pattern recognition as well as process optimization.
He proposed (2002) a new model of pattern recognition principles, witch is based on “matter cognition” instead of “matter classification” in traditional statistical pattern recognition, named point-set covering recognition or topological recognition. This new model is better closer to the function of human being, therefore it also be called Biomimetic Pattern Recognition (BPR).
In 2002 he developed a new tool named “Computational Descriptive Geometry in High Dimensional Space” for analyzing the point-set covering problem.
For hardware implementation of Neural Networks to solving the Point-set Covering Problem in high dimensional space, he created the Multi-weighted Neural Networks in 2001.
Professor Wang received the first-class award of National New Products Prize in 1964, the first-class as well as the second-class award of Scientific and Technological Achievement Prize from CAS in 1980, 1983, 1992, and 1996, the first-class award of Advanced Science and Technology Prize Beijing in 2001. He also has received twice the National Invention Prize in 1964 and again in 1996. He was given the Award of Progress in Science and Technology in 2001.