Colloquium/Seminar

YearMonth
2017 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Oct   Nov   Dec  
2016 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Oct   Nov   Dec  
2015 Jan   Feb   Mar   Apr   May   Jun   Aug   Sep   Oct   Nov   Dec  
2014 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec  
2013 Jan   Feb   Mar   Apr   May   Jun   Aug   Sep   Nov   Dec  
2012 Jan   Feb   Apr   May   Jun   Jul   Aug   Sep   Nov   Dec  
2011 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec  
2010 Jan   Feb   Mar   Apr   May   Jun   Sep   Oct   Nov   Dec  
2009 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec  
2008 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec  
2007 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec  
2006 Jan   Feb   Mar   Apr   May   Jun   Jul   Sep   Oct   Nov   Dec  
2005 Jan   Feb   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec  
2004 Jan   Feb   Mar   Apr   May   Aug   Sep   Oct   Nov   Dec  

Event(s) on February 2009


  • Monday, 9th February, 2009

    Title: Limited alphabet least squares problems in wireless communication
    Speaker: Prof. Chun, Joohwan , Scientific Computing Laboratory, Department of Electrical Engineering,, Korea Advance Institute of Science and Technology (KAIST) , Korea
    Time/Place: 11:30  -  12:30
    FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University
    Abstract: In modern wireless communication systems, multiple antennas are used in the transmitter as well as in the receiver to increase the data rate of communication. For such multiple input and multiple output (MIMO) communication systems, we seek the limited alphabet least squares solution of a complex linear system of equations. Although this problem is well-known to be NP-hard, several heuristic algorithms such as the ‘zero-forcing method’, ‘minimum-mean squares error method’, ‘V-Blast method’, ‘sphere decoding and its variations’ exist. Furthermore, preconditioning techniques such as the ‘lattice reduction method’ or ‘vector perturbation method’ can be applied either at the transmitter or receiver to improve the conditioning of the system of linear equations. In this talk, we shall review those techniques and address future possible research topics in this field.


  • Tuesday, 10th February, 2009

    Title: A supersymmetric Sawada-Kotera equation
    Speaker: Prof. Liu Qing Ping , Department of Mathematics, , China University of Mining and Technology (Beijing), China
    Time/Place: 11:30  -  12:30
    FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University
    Abstract: In this talk, first the speaker will introduce certain basic facts of soliton theory and supersymmetric integrable equations. Then the speaker will present a supersymmetric analogue for the so-called Sawada-Kotera equation and establish its integrability.


  • Monday, 16th February, 2009

    Title: 從笛卡爾到龐加萊 —— 法國數學的人文主義傳統
    Speaker: Prof. Cai Tianxin (蔡天新), Department of Mathematics, Zhejiang University, China
    Time/Place: 11:30  -  12:30
    FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University
    Abstract: Professor Cai is a professor of mathematics at Zhejiang University. He is also considered to be one of the most active young Chinese avant-garde authors with many publications on mathematics culture. 蔡天新,1963年出生於浙江台州,1978年考入山東大學控制理論專業,讀研時轉為數論方向,1987年在潘承洞院士指導下獲博士學位,現為浙江大學數學系教授、博士生導師。既主持數論方向的國家自然科學基金,也主持外國文學方向的國家社會科學基金,新近出版的《數學與人類文明》被列入國家級規劃教材,科學隨筆集《難以企及的人物》也於近年問世。 蔡天新也是一位著名的詩人和作家,有多首(篇)作品入選《中學語文》和《大學語文》新讀本,曾擔任安高詩歌獎、中國博客網大賽、日本世界俳句大賽、瑞士「中國藝術專案」的評委,他的作品被譯成英、西、法、意、德和日、韓、阿拉伯語、波斯語,世界語等18種文字,有7種外版書籍面世,先後有20次應邀參加五大洲國際詩歌節和文學節,包括香港國際文學節,並在巴黎,康橋等城市舉辦個人朗誦會。


  • Tuesday, 17th February, 2009

    Title: Analysis and synthesis of turbulent dynamic textures
    Speaker: Prof. Gabriel Peyre, CNRS and Ceremade, Universite Paris-Dauphine, France
    Time/Place: 11:30  -  12:30
    FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University
    Abstract: In this talk I will present a new grouplet transform to process locally parallel textures. This transform is well suited to the modeling of natural textures. Unlike previous transforms routinely used for texture analysis, grouplets provide a sparse representation of directional textures. Indeed, transforms like wavelets or Gabor fail to compress geometric patterns present in natural textures. The grouplet transform is implemented with a fast adaptive algorithm that extracts a geometric flow and filters recursively the texture along this flow. The resulting transformed coefficients correspond to the decomposition of the image on a multiscale tight frame adapted to the image content. The grouplet coefficients of a geometric texture are both sparse and nearly independent, which makes this representation suitable for various texture processing tasks. I will show applications to texture inpainting and dynamic texture synthesis, which both require the joint optimization of the geometric flow and the grouplet coefficients.


  • Wednesday, 18th February, 2009

    Title: JRIAM PL: Modern Mathematics and Modern Arts
    Speaker: Prof. Tianxin Cai, Jhejiang University, China
    Time/Place: 16:30  -  17:30 (Preceded by Reception at 4:00pm)
    LT1, Cha Chi-ming Science Tower, HSH Campus, Hong Kong Baptist University


  • Thursday, 19th February, 2009

    Title: Recent developments on green products
    Speaker: Prof. Arthur Huang, MINIWIZ Sustainable Energy Dev. Ltd., Taiwan
    Time/Place: 17:00  -  18:00
    FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University
    Abstract: Our company is dedicated to delivering a well-designed green product at an extremely affordable price, making hi-tech, Green power attractive and accessible to all. In this talk, we overview the history of MINIWIZ and some of our environment products¡¦ R&D. HYmini, miniNOTE, POLLI-BRICK, SOLARBULB, Bicycle Dynamo Hub Power Converter, and Solar Infrared Door Light. We will share our experience in carrying out the Far Eastern Group Pavilion in Taipei Intl. Expo 2010 project. In particular, we hope that the POLLI-BRICK, which is a recycled polymer architecture brick, can help initiate some industry-university joint research and collaboration; in terms of both modeling and numerical simulations.


  • Tuesday, 24th February, 2009

    Title: Distinguished CMIV Lecture: Multi-granular Computing-the Key to Image Classification
    Speaker: Prof. Bo Zhang, Tsinghua University, China
    Time/Place: 11:30  -  12:30
    FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University
    Abstract: The aim of the talk is to expound the basic principle and characteristic of multi-granular computing by using image classification as an example. We will present the theoretical framework of granular computing and its application to image classification. The theory is called the quotient-space based theory since we use a set of quotient spaces as a mathematical model of different grain size worlds and use the model to analyze the relationship among different worlds. In image classification, we first discuss the image representation problem. It's known that an image can be represented in computers with different granularities from fine to coarse grain-size such as pixel-based, global features, local features, etc. Each representation has its own expressiveness and robustness. But if one representation has good expressiveness then its robustness always becomes poor. Thus, the better way is to made use of the representations with different grain-sizes simultaneously, i.e., by multi-granular computing. Second, by the experimental results we show that the multi-granular computing based image classification is superior to the uni-granular one. Unfortunately, a lot of research works have been dong so far in image classification but the well-known algorithms, either based on multi-granular or uni-granular ones, are far from effectiveness. Therefore, the image classification in computer science is still an open problem. From neuroscience, many experimental results show that the information processing in human vision, for example, the information processing in human's visual cortex, supports the concept of multi-granular computing. Image classification, or in general computer vision, should learned something from those knowledge. It seems that one of future research directions that we should explore is to take full advantage of the principle of visual processing in human brain.


  • Friday, 27th February, 2009

    Title: PL: From Machine Learning to Human Innovation
    Speaker: Prof. Edward Chang, Research, Google China
    Time/Place: 16:30  -  17:30 (Preceded by Reception at 16:00.)
    WLB203, The Wing Lung Bank Building, Shaw Campus, HKBU
    Abstract: Innovation requires a prepared mind, which is a balance between ideal and necessity, between exploration and exploiting, and several other axes. In this talk, I will use machine learning theories to illustrate effective human learning principles, and use learning to contrast innovation. A few example algorithms of machine learning are used to explain how humancan learn more effectively, and a few examples in art and literature are used to illustrate human innovation. The talk will also discuss Google's culture of innovation and some of its innovative technologies.