Jie Wang


Research Assistant Professor

Department of Computational Medicine and Bioinformatics
University of Michigan, Ann Arbor



About Me

I am a Research Assistant Professor at University of Michigan working with Professor Jieping Ye. I have broad interests in optimization, machine learning, data mining, biomedical informatics etc. My current research focuses on developing novel solutions for big data. Our approach, which is called “screening”, is able to greatly reduce the data volume needed for the training phase, which leads to substantial computational savings. We have generalized our framework to many popular models, including Lasso, nonnegative Lasso, SVM, LAD, Sparse Logistic Regression, Group Lasso, Mixed-norm Regularized Regression, Fused Lasso, etc. Our screening methods, which can be integrated with any existing solvers, have been proven very promising and the efficiency can be improved by several orders of magnitude.

Research Interests

  • Optimization

  • Machine Learning and Data Mining

  • Biomedical Informatics

  • Image Processing and Pattern Recognition

  • Scaling Up Sparse Support Vector Machine by Simultaneous Feature and Sample Reduction.
    Weizhong Zhang, Bin Hong, Jieping Ye, Deng Cai, Xiaofei He, and Jie Wang.
    arXiv:1607.06996v2. [Code Download]

  • Parallel Lasso Screening for Big Data Optimization.
    Qingyang Li, Shuang Qiu, Shuiwang Ji, Jieping Ye, and Jie Wang.
    SIGKDD 2016.

  • A Multi-task Learning Formulation for Survival Analysis.
    Yan Li, Jie Wang, Jieping Ye, and Chandan Reddy.
    SIGKDD 2016.

  • Large-scale Collaborative Imaging Genetics Studies of Risk Genetic Factors for Alzheimer’s Disease Across Multiple Institutions.
    Qingyang Li, Tao Yang, Liang Zhan, Derrek Hibar, Neda Jahanshad, Yalin Wang, Jieping Ye, Paul Thompson, and Jie Wang.
    MICCAI 2016.

  • Our screening method, DPP, is highlighted by leading researchers in their new book (Section 5.10):
    Statistical Learning with Sparsity: The Lasso and Generalizations.
    Trevor Hastie, Robert Tibshirani, and Martin Wainwright

  • Multi-Layer Feature Reduction for Tree Structured Group Lasso via Hierarchical Projection. Spotlight
    Jie Wang and Jieping Ye.
    NIPS 2015.

  • Safe Screening for Multi-Task Feature Learning with Multiple Data Matrices.
    Jie Wang and Jieping Ye.
    ICML 2015.

  • Detecting genetic risk factors for Alzheimer’s disease in whole genome sequence data via Lasso screening.
    Tao Yang, Jie Wang, Qian Sun, Derrek Paul Hibar, Neda Jahanshad, Li Liu, Yalin Wang, Liang Zhan, Paul Thompson, and Jieping Ye.
    IEEE International Symposium on Biomedical Imaging, 2015.

  • DPC 2.1.1 was released, Dec 2014.
    TLFre rule for Sparse-Group Lasso is added.

  • Fused Lasso Screening Rules via the Monotonicity of Subdifferentials.
    Jie Wang, Wei Fan, and Jieping Ye.
    IEEE Transactions on Pattern Analysis and Machine Intelligence, to appear.

  • Lasso Screening Rules via Dual Polytope Projection. (Improved version of the one accepted by NIPS 2013)
    Jie Wang, Peter Wonka, and Jieping Ye.
    Journal of Machine Learning Research, 16(May):1063−1101, 2015. [Code Download]