Xingye QIAO
Research Interests:
- High Dimension Low Sample Size Data Analysis
- Statistical Machine Learning and Data Mining
- Variable Selection
- Biometrics and Bioinformatics
Education:
-
Ph.D. (2010 - expected) Statistics - University of North Carolina at Chapel Hill
Thesis Topic: Weighted Distance Weighted Discrimination and Multiscale Variable Screening
Advisors: J. S. Marron & Yufeng Liu -
M.S. (2007) Statistics - University of North Carolina at Chapel Hill
Thesis Topic: Adaptive Weighted Learning for Unbalanced Multicategory Classification
Advisor: Yufeng Liu -
B.S. (2005) Mathematics/CS - Fudan University (Shanghai, China)
Thesis Topic: Classification Methods in Facial Recognition (in Chinese)
Advisor: Zongmin Wu
Selected Publications:
See preprints
Teaching Experience:
- Instructor: STOR 151.5 Basic Statistics Summer 2009
- Instructor: STOR 151.1 Basic Statistics Fall 2008
- Teaching Assistant: STOR 155 Introductory Statistics Spring 2008
- Mentor: SAMSI/CRSC Undergraduate Workshop May 2007
- Teaching Assistant: STOR 155 Introductory Statistics Spring 2006
- Teaching Assistant: STOR 155 Introductory Statistics Fall 2005
Research:
My research interests are in the area of high-dimensional inference, with a focus on statistical machine learning and data mining methods. My research was motivated by various scientific research areas, such as genetics, drug discovery, and medical image analysis. However, the resulting methods and theory will have beneficial impacts on fields far beyond those motivating it, such as financial modeling and environmental science.
The challenges brought by high-dimensional data include how to preserve competitive performance of the methods under high-dimensional settings, and how to interpret the variables and determine which are "important". In addition, high-dimensional, low-sample size data are often counter-intuitive. My research aims to find innovative and efficient computational methods for high-dimensional data analysis, and provide thorough theoretical studies to provide statistical insight.