Chen Xu, Ph.D. November 14, 2014 The Methodology Center cux10@psu.edu The Pennsylvania State University Tel: 814-867-2512 204 E. Calder Way, Suite 400 Fax: 814-863-0000 State College, PA, 16801 http://methodology.psu.edu/people/cxu Education University of British Columbia (UBC) Vancouver, BC Ph.D. Statistics 2007 2012 Advisor: Dr. Jiahua Chen Thesis: Applications of Regularization Methods for Feature Selection in Statistical Modeling Featured courses: Asymptotic Theory, Statistics in Clinical Studies, Statistical Consulting York University (York U.) Toronto, ON M.A. Statistics 2006 2007 Featured courses: Statistical Data Mining, Applied Statistical Models, Time Series Analysis Xi an Jiaotong University (XJTU) Xi an, China B.Sc. Applied Mathematics 2002 2006 Final project was honored as the Best Undergraduate Thesis of XJTU Featured courses: Artificial Intelligence, Operations Research, Functional Analysis, Differential Geometry, Programming Softwares and Languages (R, Matlab) Research Interest I have been working on sparse modeling and statistical learning. My research interests include feature selection, regularization methods, high-dimensional regression, kernel methods, and statistical computing. Recently, I focus on developing efficient processing methods for big data, where traditional methods are less helpful due to the high computational burden. My works emphasize on both theoretical and computational aspects, which have a wide application scope in various disciplines such as genetics, biology, health science, geology, finance, and internet studies. Research Experience Research Associate (Postdoctoral Fellow) Penn State University, PA Statistical learning May 2013 Present Mentor: Dr. Runze Li Working on forward and distributive learning methods for large-scale data Methods applicable to medical imaging, marketing, and internet studies Methodologist (Ph.D. Internship) Statistics Canada, ON Analysis of survey data Jan 2010 Apr 2010 Developed sample-based variable selection method for complex surveys Methods applied to 2009 Survey on Living with Chronic Diseases in Canada
Research Assistant University of BC, BC Model selection May 2008 Nov 2012 Developed joint screening method for high-dimensional data Developed efficient thresholding algorithms for mixture models Methods applied to gene expression data, SNP data, and imrna data Research Assistant Xi an Jiaotong U., China Data mining Sep 2005 Aug 2006 Developed new approaches for visual regression and manifold reconstruction Methods applied to dimensionality reduction for imaging data Publications 1) Xu, C. and Li, R. (2014). On the Feasibility of Distributed Gaussian Kernel Regression for Big Data. Submitted to IEEE Transactions on Knowledge and Data Engineering. 2) Xu, C., Lin, S., Fang J. and Li, R. (2014). Prediction-based Termination Rule for Greedy Learning with Massive Data. Submitted to Statistica Sinica. 3) Xu, C. and Chen, J. (2014). The Sparse MLE for Ultra-high-dimensional Feature Screening. Journal of the American Statistical Association, 109, 1257-1269. 4) Lin, S., Xu, C., Zeng, J. and Fang, J. (2014). Does the Generalization Capability of L q Regularization Depend on the Choice of q? Revised for Constructive Approximation. 5) Xu, C. and Chen, J. (2013). A Thresholding Algorithm for Order Selection in Finite Mixture Models. Communications in Statistics, 44, 433-453. 6) Xu, C., Chen, J. and Mantel, H. (2013). Pseudo-likelihood-based Bayesian Information Criterion in Analysis of Survey Data. Survey Methodology, 39, 303-321. 7) Xu, C., Peng, Z. and Jing, W. (2011). Sparse Kernel Logistic Regression Based on L.5 Regularization. Science of China, 53, 1-17. 8) Xu, C., Chen, J. and Mantel, H. (2010). Smoothly Clipped Absolute Deviation in Analysis of Survey Data. Proceedings of the Survey Methods Section, Quebec city, Statistical Society of Canada. (Awarded the Best Student Paper in Survey Method Section of SSC meeting 2010) 9) Meng, D., Xu, C. and Xu, Z. (2010). A New Manifold Reconstruction Method based on Isomap. Chinese Journal of Computers, 33, 546-555. 10) Meng, D., Xu, C. and Jing, W. (2005). A New Approach for Regression: Visual Regression Approach. Lecture Notes in Computer Science, 3801, 139-144. Working Papers & Manuscripts 1) Xu, C. and Li, R. (2015). The Penalized Likelihood Ratio Test for High-dimensional Models. In progress. 2) Xu. C. and Zou, B. (2015). The Efficient Greedy Learning for Big Data. Manuscript.
3) Li, R., Huang, Y., Wang, L. and Xu, C. (2015). Projection Test for High-dimentional Mean Vectors with Optimal Direction. Manuscript. 4) Xu. C. and Lin, S. (2015). Learning through Deterministic Assignments of Hidden Parameters. Manuscript. Awards & Honours International Tuition Fee Scholarship (UBC )....................... 2011-2012 Best Student Paper Award Presented at SSC Meeting (Survey Methods Section, SSC )... 2010 Ph.D. Research Fellowships (MATICS and Statistics Canada)................. 2010 Graduate Entrance Scholarship (UBC ).............................. 2007 Ph.D. Tuition Fee Award (UBC )............................ 2007 2010 International Entrance Scholarship (York U.)........................... 2006 The Best Undergraduate Thesis of the University (XJTU )................... 2006 The 2nd-Class Academic Scholarship for Students (XJTU ).............. 2004 2005 Student Leadership Award (XJTU )........................... 2003 2005 Merit Student Award (XJTU ).............................. 2002 2004 Teaching Experience Survey Sampling (STAT 344) University of BC Teaching Assistant Winter 2008 Designed and led weekly labs; Guided students using software R. Created and marked assignments/exams; Held weekly office hours. Received a 4.7 student evaluation (on a 1 to 5 scale). Statistical Methods (STAT 203) University of BC Teaching Assistant Fall 2007 Led weekly tutorials to guide students through difficult problems. Received a 4.4 student evaluation (on a 1 to 5 scale). Elementary Statistics II (MATH 2570) York University Teaching Assistant Winter 2007 Led weekly group tutorials and reviewed course materials. Fundamentals of Mathematics (MATH 1710) York University Teaching Assistant Fall 2006 Provided one-on-one and group tutorials to students in first and second year math courses. Online tutorials for international long-distance students. Presentations & Posters Efficient greedy learning for massive data Summer Workshop Xi an Jiaotong University, Xi an, China (Jul 2014)
The sparse MLE for ultra-high dimensional feature screening Research Seminar University of Waterloo, Waterloo, ON (Jan 2014) Soft thresholding-based screening for ultra-high dimensional feature spaces Joint Statistical Meetings San Diego, CA (Aug 2012) International Workshop on Perspectives on High-dimensional Data Analysis Fields Institute, Toronto, ON (Jun 2011). Detecting the changes in quality of lumber via semi-parametric density ratio model Sessional Group Meeting, FPInnovations-NSERC Collaborative Research and Development Grant FPInnovations, Vancouver, BC (May 2011). Penalized likelihood methods for variable selection in analysis of survey data Survey Method Section, the SSC Annual Meeting, Quebec City, QB (May 2010). Research Internship Report Statistics Canada, Ottawa, ON (Apr 2010). Adaptive regression by mixing Seminar on Information Theory and Pattern Recognition Institute for Information and System Sciences, Xi an Jiaotong University, Xi an, China (Jul 2008). A Maximum Likelihood Methodology for Clusterwise Linear Regression Graduate Student Seminar University of BC, Vancouver, BC (Oct 2009). Research meeting York University, Toronto, ON (May 2007). Academic Service Guest Editor Special Issue on Machine Learning for Medical Imaging, Neurocomputing (2015). Article Reviewer Annals of Statistics Journal of American Statistic Association Canadian Journal of Statistics Journal of Multivariate Analysis Conference Volunteer The C. R. & Bhargavi Rao Prize Conference, State College, PA (Oct 2013). 2010 Joint Statistical Meetings, Vancouver, BC (Aug 2010). The 2nd Pao-Lu Hsu Conference of Machine Learning, Xi an, China (Jul 2010). Annual Meeting of Statistic Society of Canada, Vancouver, BC (Jun 2009).
Statistical Consulting and Other Projects Xu, C. and Chen, J. (2012). Analysis of mirna Data based on Logistic Model in Oral Cancer Study. Consulting report. To Faculty of Dentistry, University of BC, Vancouver. Applied the screening-based regularization method to identify the influential mirnas to the oral cancer patients. Xu, C. (2011). Lumber Quality Assessment via Semi-parametric Density Ratio Model. Consulting report. To FPInnovations, Vancouver. Developed the semi-parametric density ratio model for assessing the quality of lumber. Xu, C. (2010). The Consistency of SCAD for Variable Selection in Complex Surveys. Internship report. To Statistics Canada, Ottawa. Proposed the sample-based regularization approach for variable selection in complex surveys. Xu, C. (2009). Rates of Vocalization of a Beluga Calf. Consulting report. To Vancouver Aquarium, Vancouver. Designed the regression methods to detect the factors that can stimulate the vocal development of a beluga. Xu, C. (2009). Medical Terms - Early Twentieth Century Chinese History. Consulting report. To Department of History, University of BC, Vancouver. Designed the clustering methods to learn about patterns of relationships and activities through the study of collective biography of people. Xu, C. (2008). Surface Area Determination of Waste Rock Particles. Consulting report. To Norman B. Keevil Institute of Mining Engineering, Vancouver. Designed the testings methods to compared the performances of different measuring approaches on small particles. Calculated the sample size needed for estimating the population mean with a given accuracy level. Xu, C. and Shen, T. (2008). The Effects of Environmental Variability on Evolution of Mean and Variance in Wing Size in Drosophila. Consulting report. To Department of Zoology, University of BC, Vancouver. Designed the testings methods to evaluate the environmental effects on the growth of fruit flies.