Papers

  • Shape-constrained Estimation

    We study the estimation of monotone functions, convex/concave functions, and functions with higher-order shape constraints. These functions occur naturally, admit efficient estimation algorithms, and exhibit interesting statistical properties.

    Faithful Variable Screening for High-dimensional Convex Regression.
    Min Xu, Minhua Chen, John Lafferty
    arXiv preprint 2014
    Mini-Abstract: We show that, for the purpose of identifying irrelevant variables in high-dimensional estimation, it is in some sense sufficient to approximate a general convex function with an additive convex function.

    Learning High-Dimensional Concave Utility Functions for Discrete Choice Model
    Yuxue Qi, Min Xu, John Lafferty
    preprint
  • Representing Documents Through Their Users
    Khalid El-Arini, Min Xu, Emily Fox, Carlos Guestrin
    KDD 2013
  • Estimation Bias in Multi-armed Bandit Algorithms for Search Advertising
    Min Xu, Tao Qin, Tie-Yan Liu
    NIPS 2013
  • Conditional Sparse Coding and Multivariate Regression for Grouped Data.
    Min Xu, John Lafferty
    ICML (conference) 2012 version Appendix
  • Guarantees for Spectral Clustering

    We bound the mis-clustering rate for spectral clustering under various settings and assumptions.

    Robust Active Spectral Clustering.
    Akshay Krishnamurthy, Sivaraman Balakrishnan, Min Xu, Aarti Singh
    ICML (conference) 2012 version.

    Noise Thresholds for Spectral Clustering.
    Sivaraman Balakrishnan, Min Xu, Akshay Krishnamurthy, Aarti Singh
    NIPS (conference) 2011 version.
  • High-Dimensional Covariance Estimation Based on Gaussian Graphical Models.
    Shuheng Zhou, Phillip Rutimann, Min Xu, Peter Buhlmann
    JMLR (journal) 2011 version
  • Forest Density Estimator.
    Han Liu, Min Xu, Haijie (Jay) Gu, John Lafferty, Larry Wasserman, Anupam Gupta
    COLT (conference) 2010 version. JMLR (journal) 2011 version.

Software