Canyi Lu (卢参义)


I am now the final year phd student with Department of Electrical and Computer Engineering at National University of Singaporedirected by Profs. Shuicheng Yan and Zhouchen Lin (Peking University). Before that, I obtained my master degree from the University of Science and Technology of China (USTC).

canyilu AT gmail.com





Current Research Interests
  • Statistical learning of structured sparsity: bounds for exact and stable recovery for models and solvers
    Sparse structures of interest: sparse vector, low-rank matrix, block diagonal matrix and low-rank tensor
  • Numerical optimization, including convex/nonconvex, constrained/non-constrained, stochastic methods
  • Applications in PR and CV: subspace clustering, multi-view learning, signal recovery...              

Research Highlights: 3 frameworks and 1 generalization

  •  Unified theory of block diagonal matrix learning for subspace clustering
   
1. Robust and Efficient Subspace Segmentation via Least Squares Regression. C. Lu, H. Min, Z. Zhao, L. Zhu, D. Huang, S. Yan. ECCV, 2012.
2. Subspace Clustering by Block Diagonal Representation, C. Lu, J. Feng, Z. Lin, S. Yan. TPAMI, 2017. Major revision.

  • Unified framework of Alternating Direction Method of Multipliers (ADMM)
    
1. A Unified Alternating Direction Method of Multipliers by Majorization Minimization. C. Lu, J. Feng, Z. Lin, S. Yan. TPAMI. 2017
2. LibADMM: A Library of Alternating Direction Method of Multipliers for Compressive Sensing. C. Lu. 2016. [GitHub Link]

  •  Unified solvers for nonconvex nonsmooth low-rank minimization
     
1. Nonconvex Nonsmooth Low-Rank Minimization via Iteratively Reweighted Nuclear Norm. C. Lu, J. Tang, S. Yan, Z. Lin. TIP. 2016.
2. Generalized Singular Value Thresholding. C. Lu, C. Zhu, C. Xu, S. Yan, Z. Lin. AAAI. 2015.

  •  Low-rank tensor completion and decomposition: theory and applications
   

1. Tensor Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Tensors via Convex Optimization. C. Lu, J. Feng, Y. Chen, W. Liu, Z. Lin, S. Yan. CVPR. 2016.
2. Tensor Recovery via Convex Optimization. C. Lu, J. Feng, Z. Lin, S. Yan. In submission. 2016


Selected Awards
  • Stars of Tomorrow (Award of Excellent Intern), Microsoft Research Asia, 2015
  • Microsoft Research Asia Fellowship, 2014 (12 Phd students in Asia). Link1Link2
  • ICCV13 Travel Grant, 2013
  • Outstanding Postgraduate Award, 2011
  • Outstanding Graduate Award, 2009
  • Excellent Bachelor Dissertation Award, 2009
  • National Scholarship (2 times), 2006, 2007
  • National Scholarship for Encouragement, 2007
  • Excellent Student Scholarship (7 times), 2005-2009

Preprints



          Tensor Recovery via Convex Optimization
               Canyi LuJiashi Feng, Shuicheng Yan, and Zhouchen Lin
                In submission. 2016

     Subspace Clustering by Block Diagonal Representation
               Canyi LuJiashi Feng, Zhouchen Lin and Shuicheng Yan
                In submission to TPAMI, major revision, 2017

    o Accelerated Stochastic Mirror Descent Algorithms for Composite Non-strongly Convex Optimization
     Le Thi Khanh Hien, Canyi LuHuan Xu, Jiashi Feng
     [arXiv]

Optimized Projections for Compressed Sensing via Direct Mutual Coherence Minimization
     
Canyi Lu, Huan Li, Zhouchen Lin
     In submission. 2016
     
[arXiv][BibTex]


Projection onto the Capped Simplex
     
Weiran Wang, Canyi Lu
     [PDF][arXiv][Matlab and C++ implementation][BibTex]

Selected Publications

A Unified Alternating Direction Method of Multipliers by Majorization Minimization

     Canyi Lu, Zhouchen LinShuicheng Yan
      Accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI, 2017

Robust and Efficient Subspace Segmentation via Least Squares Regression
    Can-Yi Lu, Hai Min, Zhong-Qiu Zhao, Lin Zhu, De-Shuang Huang, Shuicheng Yan
     European Conference on Computer Vision (ECCV) (7) 2012: 347-360
     [PDF][arXiv][Code][BibTex]

     In Table 1, the nuclear norm does not satisfy the EBD condition

Tensor Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Tensors via Convex Optimization
     Canyi Lu, Jiashi Feng, Yudong Chen, Wei Liu, Zhouchen Lin, Shuicheng Yan
    IEEE International Conference on Computer Vision and Pattern Recognition (CVPR). 2016
     [PDF][arXiv][LibADMM Toolbox][BibTex]

o Generalized Singular Value Thresholding
    Canyi LuChangbo Zhu, Chunyan Xu, Shuicheng Yan, Zhouchen Lin
     Proceedings of the AAAI Conference on Artificial Intelligence (AAAI). 2015 
    [PDF][arXiv][Code][BibTex]

Nonconvex Nonsmooth Low-Rank Minimization via Iteratively Reweighted Nuclear Norm
     Canyi LuJinhui Tang, Shuicheng Yan, Zhouchen Lin
     IEEE Transactions on Image Processing (TIP). 2016 (longer version of CVPR 2014 paper)
     [PDF][arXiv][BibTex][Code of CVPR version]

o Fast Proximal Linearized Alternating Direction Method of Multiplier with Parallel Splitting
    Canyi LuHuan Li, Zhouchen Lin, Shuicheng Yan
    Proceedings of the AAAI Conference on Artificial Intelligence (AAAI). 2016
    [PDF][arXiv][BibTex]

Generalized Nonconvex Nonsmooth Low-Rank Minimization
     Canyi LuJinhui Tang, Shuicheng Yan, Zhouchen Lin
     IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) 2014: 4130-4137
     [PDF][arXiv][Code][BibTex]

Convex Sparse Spectral Clustering: Single-view to Multi-view
     Canyi LuShuicheng Yan, Zhouchen Lin
     IEEE Transactions on Image Processing (TIP). 2016
     [PDF][arXiv][Code][BibTex]

Correlation Adaptive Subspace Segmentation by Trace Lasso
    Canyi Lu, Jiashi Feng, Zhouchen Lin, Shuicheng Yan
     IEEE International Conference on Computer Vision (ICCV) 2013: 1345-1352
    [PDF][arXiv][BibTex]

Smoothed Low Rank and Sparse Matrix Recovery by Iteratively Reweighted Least Squares Minimization
     Canyi Lu, Zhouchen Lin, Shuicheng Yan
     IEEE Transactions on Image Processing (TIP) 24(2): 646-654 (2015)

Correntropy Induced L2 Graph for Robust Subspace Clustering
     Canyi Lu, Jin­hui Tang, Min Lin, Liang Lin, Shuicheng Yan, Zhouchen Lin
     IEEE International Conference on Computer Vision (ICCV) 2013: 1801-1808
    [PDF][arXiv][BibTex]

Proximal Iteratively Reweighted Algorithm with Multiple Splitting for Nonconvex Sparsity Optimization
   Canyi Lu, Yunchao Wei, Zhouchen Lin, Shuicheng Yan    
     Proceedings of the AAAI Conference on Artificial Intelligence (AAAI). 2014
     [PDF][arXiv][BibTex]

Collaborative Completion of Transcription Factor Binding Profiles via Local Sensitive Unified Embedding
      
Lin Zhu, Wei-Li Guo, Canyi Lu, De-Shuang Huang
     IEEE Transactions on NanoBioscience. 2016

Connections Between Nuclear Norm and Frobenius Norm Based Representation
      
Xi Peng, Canyi Lu, Zhang Yi, Huajin Tang
     IEEE Transactions on Neural Networks and Learning Systems (TNNLS). 2016

o Cross-​modal Re­trieval with CNN Vi­su­al Fea­tures: A New Base­line
    Yun­chao Wei, Yao Zhao, Canyi Lu, Shikui Wei, Luoqi Liu, Zhen­feng Zhu, Shuicheng Yan
     IEEE Trans­ac­tions on Cy­ber­net­ics (TCYB). 2016

Multi-Loss Regularized Deep Neural Network
    Chunyan Xu, Canyi Lu, Xiaodan Liang, Junbin Gao, Tianjiang Wang, Shuicheng Yan
     IEEE Transactions on Circuits and Systems for Video Technology (TCSVT). 2015
     [PDF]

Imputation of ChIP-seq Datasets via Low Rank Convex Co-Embedding 
    Lin Zhu, Wei-Li Guo, De-Shuang Huang, Canyi Lu
    IEEE International Conference on Bioinformatics and Biomedicine (BIBM). 2015

Discriminative Analysis for Symmetric Positive Definite Matrices on Lie Groups 
    Chunyan Xu, Canyi Lu, Junbin Gao, Tianjiang Wang, Shuicheng Yan
      IEEE Transactions on Circuits and Systems for Video Technology (TCSVT25(10): 1576-1585 (2015) 
    [PDF][BibTex]

Facial Analysis with Lie Group Kernel
    Chunyan Xu, Canyi Lu, Junbin Gao, Tianjiang Wang, Shuicheng Yan
      IEEE Transactions on Circuits and Systems for Video Technology (TCSVT 25(7): 1140-1150 (2015)
    [PDF][BibTex]

    Adaptive Nonparametric Image Parsing
          Tam Nguyen, Canyi Lu, Jose Sepulveda, Shuicheng Yan
              IEEE Transactions on Circuits and Systems for Video Technology (TCSVT25(10): 1565-1575 (2015)
               [PDF][BibTex]

Efficient k-Support Matrix Pursuit
    Hanjiang Lai, Yan Pan, Canyi Lu, Yong Tang, Shuicheng Yan
     European Conference on Computer Vision (ECCV(2) 2014: 617-631 
    [PDF][BibTex]

Face Recognition via Weighted Sparse Representation

        Can-Yi Lu, Hai Min, Jie Gui, Lin Zhu, Ying-Ke Lei
        Journal of Visual Communication and Image Representation. 24(2): 111-116 (2013)
          
[PDF][Code][BibTex][Top 25 Most Cited JVCI Article Published Since 2010]


Optimized Projections for Sparse Representation based Classification 

         Can-Yi Lu, De-Shuang Huang
         Neurocomputing 113: 213-219 (2013)
       
  
[PDF][Code][BibTex]

 


Academic Services

1. Journal Reviewer for
  • Journal of Machine Learning Research (JMLR)
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
  • Machine Learning
  • IEEE Transactions on Knowledge and Data Engineering (TKDE)
  • IEEE Transactions on Image Processing (TIP)
  • IEEE Transactions on Signal Processing (TSP)
  • Computer Vision and Image Understanding (CVIU)
  • ACM Transactions on Multimedia Computing, Communications and Applications (TOMM)
  • IEEE Transactions on Multimedia (TMM)
  • Neural Networks
  • Neurocomputing
  • Neural Computing and Applications (NCAA)
  • Pattern Analysis and Applications (PAA)
  • Journal of Visual Communication and Image Representation (JVCI)
  • The Visual Computer
  • IEEE Signal Processing Letters (SPL)
  • Circuits, Systems, and Signal Processing (CSSP)
2. Conference Program Committee/Reviewer for
  • IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016, 2017
  • International Conference on Computer Visions (ICCV) 2015, 2017
  • European Conference on Computer Vision(ECCV) 2016
  • AAAI Conference on Artificial Intelligence (AAAI) 2016, 2017
  • International Joint Conference on Artificial Intelligence (IJCAI2016
  • International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2016
  • Asian Conference on Computer Vision (ACCV) 2016

Locations of visitors to this page


 Vistiors  
 Visits    asp hit counter