Semiparametric models are useful to determine features that are invariant to certain conditions, such as illumination invariant features in object recognition problems. In this article we propose a general semiparametric model, which is called the Generalized Multiparameter Likelihood Models, to adapt with various applications, an simple and efficient estimation procedure, and an automatic procedure to identify features that are invariant to certain conditions. This is a co-work with Prof. Ming-Yen Cheng and Dr.Wenyang Zhang.
Lu-Hung Chen is an assistant professor in statistics at National Chung Hsing University. He received his Ph.D. degree in mathematics and M.S. degree in computer science and information engineering from National Taiwan university. His current research interests include nonparametric/semiparametric modeling, high dimensional data analysis, machine learning, and image processing.