学术交流报告通知

报 告 人:华南理工大学 谭明奎教授 

报告时间:2017213日上午9:00

报告地点:219会议室

报告题目:

MPGL: An Efficient Matching Pursuit Method for Generalized LASSO

报告人简介:

谭明奎,现年33岁,湖北巴东人,分别在2006年和2009年于湖南大学获得环境工程学士和模式识别与智能系统硕士学位,在201410月南洋理工大学获得计算机科学博士学位。自20146月到20169月,在阿德莱德大学沈春华教授组从事博士后研究,并于今年9月回国工作,获聘为华南理工大学教授/博导,目前主要研究方向为机器学习和机器视觉。截止今年11月,发表一作期刊论文6, 包括Journal of Machine Learning Research(1,CCF-A)IEEE Trans 4篇;一作顶级会议论文9篇,包括ICML(2篇,CCF-A)CVPR(2篇,CCF-A)AAAI(3篇,CCF-A)IJCAI(1篇,CCF-A)CIKM(1篇,CCF-B)。截止201611月,被SCI他人引用160次,h指数12

报告内容简介:

Unlike traditional LASSO enforcing sparsity on the variables, Generalized LASSO (GL) enforces sparsity on a linear transformation of the variables, gaining flexibility and success in many applications. However, many existing GL algorithms do not scale up to high-dimensional problems, and/or only work well for a specific choice of the transformation. We propose an efficient Matching Pursuit Generalized LASSO (MPGL) method, which overcomes these issues, and is guaranteed to converge to a global optimum.

We formulate the GL problem as a convex quadratic constrained linear programming (QCLP) problem and tailor-make a cutting plane method. More specifically, our MPGL iteratively activates a subset of nonzero elements of the transformed variables, and solves a subproblem only involving the activated elements thus gaining significant speed-up. Moreover, MPGL is less sensitive to the choice of the trade-off hyper-parameter between data fitting and regularization, and mitigates the long-standing hyper-parameter tuning issue in many existing methods. Experiments demonstrate the superior efficiency and accuracy of the proposed method over the state-of-the-arts in both classification and image processing tasks.