Title: | A unified framework for variance estimation in nonparametric regression |
Speaker: | Dr. TONG Tiejun, Department of Mathematics, Hong Kong Baptist University , Hong Kong |
Time/Place: | 11:30 - 12:30 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | Nonparametric regression models are very attractive in statistical data analysis and have been extensively studied in the past several decades. There is a large body of literature on the estimation of the mean function including the kernel estimators, the local linear estimators and the smoothing spline estimators. Apart from the mean function, the estimation of the residual variance is also been recognized as an equally important problem. In this talk, I will first introduce some difference-based methods for estimating the residual variance in nonparametric regression. I will then introduce some remaining challenges and some recent advances in this area. In particular, I will introduce a unified framework for variance estimation that combines the higher-order difference sequence and the linear regression technique systematically. It turns out that the unified framework generates a very large family of estimators that include most existing estimators as special cases. Finally, I will conclude the talk by some simulation studies that demonstrate the usefulness of the proposed framework. |
Title: | Global asymptotic stability and Lyapunov functionals of epidemic models with time delay |
Speaker: | Dr. Xiang-Sheng Wang, Department of Mathematics, Southeast Missouri State University, USA |
Time/Place: | 11:00 - 12:00 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | In this talk, we will start with a simple and fundamental epidemic model to illustrate the technique of Lyapunov functional in global stability analysis of equilibrium. Next, we introduce time delays to the model and state a notorious open problem related to this delay epidemic model. We will discuss on the main difficulties and finally present some recent progress in approaching this open problem. This talk is based on the thesis work of my student Calah Paulhus at Southeast Missouri State University. |
Title: | L_inf norm in Image Processing |
Speaker: | Dr. WOO Hyenkyun, School of Computational Sciences, Korea Institute for Advanced Study, Korea |
Time/Place: | 11:00 - 12:00 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | In this talk, we will show the usefulness of L_inf norm in speckle reduction problems in SAR(synthetic aperture radar) and robust nonnegative matrix factorization. 1. Speckle (multiplicative noise) naturally appear in various coherent imaging systems, such as SAR. Due to the strong interference phenomena in coherent imaging systems, it is hard to identify the valuable objects from the captured noise signal. In this talk, we introduce framework for total variation based speckle reduction problems. The framework is based on m-th root transformation and alternating minimization algorithm. 2. In the second part of this talk, we introduce L_inf norm based new low rank promoting regularization framework, i.e., asymmetric soft regularization framework for robust nonnegative matrix factorization (NMF). The main advantage of the proposed soft regularization is that it is less sensitive to the rank selecting regularization parameters since we use soft regularization instead of using the conventional hard constraints such as nuclear norm, gamma2-norm, or rank itself. |
We organize conferences and workshops every year. Hope we can see you in future.
Learn MoreProf. M. Cheng, Dr. Y. S. Hon, Dr. K. F. Lam, Prof. L. Ling, Dr. T. Tong and Prof. L. Zhu have been awarded research grants by Hong Kong Research Grant Council (RGC) — congratulations!
Learn MoreFollow HKBU Math