Gyemin Lee

Gyemin Lee received the B.S. in Electrical Engineering from Seoul National University, Seoul, Korea in 2001 and the M.S. and Ph.D. in Electrical Engineering: Systems from the University of Michigan in 2007 and 2011, respectively. He is currently a post-doctoral research fellow in Computer Science and Engineering at the University of Michigan. His research interests include machine learning, pattern recognition, optimization and statistical signal processing.

Ph.D., Post Doctoral Research Fellow
Electrical Engineering and Computer Science
University of Michigan

Contact information

Office: 4817 CSE
Email:

Research

Research advisor: Prof. Clayton Scott

Research interests: Machine learning, Pattern recognition, Statistical learning theory, Statistical file matching, Optimization, Compressive sensing, Statistical signal processing.

Papers

Submitted Papers

• G. Lee and C. Scott, "EM algorithms for multivariate Gaussian mixture models with truncated and censored data,"
  accepted to Computational Statistics and Data Analysis
  Tech. Rep. CSPL-401, Communications and Signal Processing Laboratory, University of Michigan. [pdf | code]

Journal Papers

• G. Lee, W. Finn and C. Scott, "Statistical file matching of flow cytometry data,"
  Journal of Biomedical Informatics, vol. 44, no. 4, pp. 663-676, 2011. [pdf | link | code]
• G. Blanchard, G. Lee and C. Scott, "Semi-Supervised Novelty Detection,"
  Journal of Machine Learning Research, vol. 11, pp. 2973-3009, 2010. [pdf]
• G. Lee and C. Scott, "Nested support vector machines,"
  IEEE Trans. Signal Processing, vol. 58, no. 3, 1648-1660, 2010 [pdf | code]

Conference Papers

• G. Blanchard, G. Lee and C. Scott, "Generalizing from Several Related Classification Tasks to a New Unlabeled Sample,"
  accepted to Advances in Neural Information Processing Systems (NIPS), 2011.
• G. Lee, L. Stoolman and C. Scott, "Transfer Learning for Automatic Gating of Flow Cytometry Data,"
  accepted for oral presentation at the ICML 2011 Workshop on Unsupervised and Transfer Learning, 2011.
  Pascal2 best student paper award.
  Also Tech. Rep. CSPL-402, Communications and Signal Processing Laboratory, University of Michigan. [pdf]
• G. Lee and C. Scott, "Nested support vector machines,"
  Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2008), Las Vegas, 2008. [pdf | code]
• G. Lee and C. Scott, "The one class support vector machine solution path,"
  Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2007), vol. 2, II-521--II-524, Honolulu, USA, April 2007. [pdf | movies | code]

Codes

• TCEM
  
  EM algorithm for fitting a multivariate Gaussian mixture model with truncated and censored data.

  G. Lee and C. Scott, "EM algorithms for multivariate Gaussian mixture models with truncated and censored data"

• Cluster nearest neighbor
  
  algorithm for file matching, and associated EM algorithm for fitting a mixture of PPCA model with missing attributes.

  G. Lee, W. Finn and C. Scott, "Statistical file matching of flow cytometry data"

• Nested support vector machines
  
  Matlab code to generate cost-sensitive and one-class SVMs that are properly nested (unlike standard SVMS) as the cost-asymmetry or density level parameter is varied. The solution paths are piecewise linear with a user-selected number of breakpoints.

  G. Lee and C. Scott, "Nested support vector machines"

• SVM path algorithms
  
  Matlab code to generate solution paths for the cost-sensitive SVM with varying cost-asymmetry, and the one-class SVM with varying density level parameter. The algorithms were inspired by the path algorithm of Hastie et al., which varies a regularization parameter, and were implemented for comparison with the nested SVM code above. The CS-SVM algorithm is different from the one developed by Bach et al. in that we capture the cost asymmetry in a single parameter. The OC-SVM path algorithm was detailed here:

  G. Lee and C. Scott, "The one class support vector machine solution path"

last updated: Mar 13, 2012