|2018||Jan Feb Mar Apr May Jun Jul|
|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 June 2008
- Monday, 2nd June, 2008
Title: Ultrasound Breast-Cancer Imaging with a Ring Transducer Array Speaker: Dr. Lianjie Huang, Los Alamos National Laboratory, USA Time/Place: 11:30 - 12:30
RRS905 Conference Room
Abstract: One in seven women in the United States will develop breast cancer, and the breast cancer death rate has changed little since 1930s. Early cancer detection is the key to reducing cancer mortality. Clinical ultrasound tomography is a new, promising tool to detect breast cancer at its earliest stage. We have developed a number of ultrasound reflection and transmission tomography algorithms for high-resolution breast-cancer imaging, including a super-resolution imaging method. In collaboration with Karmanos Cancer Institute, we have applied our advanced ultrasound tomography algorithms to in vitro and clinical ultrasound breast data acquired using a novel prototype ultrasound scanner with a ring transducer array. I will present ultrasound tomography imaging results of in vitro and in vivo applications. The results demonstrate that high-resolution ultrasound tomography has the potential to significantly enhance our capability of breast cancer detection and diagnosis.
- Tuesday, 17th June, 2008
Title: An MM Algorithm for Multicategory Vertex Discriminant Analysis Speaker: Dr. Tongtong Wu, MATH, School of Public Health, University of Maryland,College Park, USA Time/Place: 11:30 - 12:30
LT2, Cha Chi-Ming Science Tower, HSH Campus, Hong Kong Baptist University
Abstract: This talk introduces a new method of supervised learning based on linear discrimination among the vertices of a regular simplex in Euclidean space. Each vertex represents a different category. Discrimination is phrased as a regression problem involving $epsilon$-insensitive residuals and a quadratic penalty on the coefficients of the linear predictors. The objective function can by minimized by a primal MM (majorization-minimization) algorithm that (a) relies on quadratic majorization and iteratively reweighted least squares, (b) is simpler to program than algorithms that pass to the dual of the original optimization problem, and (c) can be accelerated by step doubling. Limited comparisons on real and simulated data suggest that the MM algorithm is competitive in statistical accuracy and computational speed with the best currently available algorithms for discriminant analysis.
- Thursday, 19th June, 2008
Title: New Territory for Low Discrepancy Designs Speaker: Prof. Fred J. Hickernell, Department of Applied Mathematics, Illinois Institute of Technology, USA Time/Place: 11:00 - 12:00
RRS905 Conference Room
Abstract: Low discrepancy points yield efficient algorithms for approximating high dimensional integrals. For some applications, such as pricing continuously monitored financial derivatives (options), the nominal dimension is not only large, but infinite. Here, the price of the derivative is the expected value of a functional of a continuous time Brownian motion. Although the continuous time Brownian motion is infinite-dimensional, practical computations are limited to a d-dimensional truncated Karhunen-Loeve expansion. It is demonstrated how the dimension, d, should vary with the sample size, n, to obtain an efficient approximation using low discrepancy designs. Prof. Fang Kai-Tai has pioneered and developed the application of low discrepancy points as designs for laboratory and computer experiments. As an example of a computer experiment, consider the design of a nuclear reactor, where there are tens of parameters to be chosen by the nuclear engineer. A computer simulation of the underlying reactor physics for a particular choice of parameters might require a day or more of parallel computing time. Searching for the best parameter values is facilitated by optimizing a surrogate, i.e., an approximation of the reactor response as a function of design parameter values based on the limited number of computer simulation runs available. This is essentially the problem of approximating a function of many variables. Unlike results from classical spectral and meshfree methods on regular grids, it is shown that low discrepancy designs lead to convergence rates that do not suffer the curse of dimensionality. The ongoing work described in this talk is a collaboration with a number of scholars including past and present HKBU students: Leung King-Tai, Liu Kwong-Ip, Niu Ben, and Zeng Xiaoyan.
- Friday, 27th June, 2008
Title: Numerical Algorithms for Modeling Multi-Physics Phenomena Speaker: Prof. CAI Wei, MATH, University of North Carolina at Charlotte, USA Time/Place: 11:00 - 12:00
SCT909, Cha-Chi-ming Science Tower, HSH Campus
Abstract: In this talk, we will highlight the development of numerical algorithms in the following areas where multiscale physical phenomena pose much challenges in achieving computational efficiency.  Hybrid explicit/implicit salvation model for long range electrostatic interactions in protein folding;  Adaptive conservative cell average spectral element methods of Wigner equations for multi-physics models of electron transport in Nano-devices;  High frequency electromagnetic scattering in dispersive media in Phase shift Mask in nanoscale X-ray Lithography;  Modeling of contact and non-Newtonian fluid mechanics with random rough surface in Chemical Mechanical Polishing (CMP) in VLSI chip design.