浙江工业大学物理学院
首页 | 旧版网站(内网) | 联系我们 | 浙江工业大学
 
 所在位置：首页 > 博学堂讲座
博学堂讲座
Efficient Approaches for Two Big Data Matrix Optimization Problems (第110讲)
浏览量：1520    发布时间：2015-06-09 08:54:17

报告题目:Efficient Approaches for Two Big Data Matrix Optimization Problems

报告人：吕召松 教授

报告时间：下午3:30-4:30

报告地点：博C107

题目：Efficient Approaches for Two Big Data Matrix Optimization Problems
报告人：吕召松教授（Simon Fraser University, Canada）
地点：屏峰校区博C107
时间：2015年6月10日（周三）下午3:30-4:30
摘要：In the first part of this talk, we consider low rank matrix completion problem, which has wide applications such as collaborative filtering, image inpainting and Microarray data imputation. We present an efficient and scalable algorithm for matrix completion. In each iteration, we pursue a rank-one matrix basis generated by the top singular vector pair of the current approximation residual and update the weights for all rank-one matrices obtained up to the current iteration. We further propose a novel weight updating rule to reduce the time and storage complexity, making the proposed algorithm scalable to large matrices. We establish a linear rate of convergence for the algorithm. Numerical experiments demonstrate that our algorithm is much more efficient than the state-of-the-art algorithms while achieving similar or better prediction performance.
 
In the second part we consider the problem of estimating multiple graphical models simultaneously using the fused lasso penalty, which encourages adjacent graphs to share similar structures. One important application of this problem is for the analysis of brain networks of Alzheimer's disease. We establish a necessary and sufficient condition for the graphs to be decomposable. As a consequence, a simple but effective screening rule is proposed, which decomposes large graphs into small subgraphs and dramatically reduces the overall computational cost. Numerical experiments demonstrate the effectiveness and efficiency of our proposed approach.
 
报告人简介：吕召松，加拿大西蒙弗雷泽大学终身教授。2005年获美国佐治亚理工学院运筹学博士学位。主要研究兴趣在大规模连续优化问题的理论和算法，及其在数据挖掘、信号处理等领域的应用。目前担任SIAM Journal on Optimization, Big Data and Information Analytics杂志副主编，主持加拿大自然科学与工程技术研究理事会（NSERC）基金3项，发表学术论文40余篇，其中20余篇发表在运筹与优化国际顶级期刊Mathematical Programming和SIAM Journal on Optimization等上。
 
博学堂讲座
Efficient Approaches for Two Big Data Matrix Optimization Problems (第110讲)
浏览量：1520    发布时间：2015-06-09 08:54:17

报告题目:Efficient Approaches for Two Big Data Matrix Optimization Problems

报告人：吕召松 教授

报告时间：下午3:30-4:30

报告地点：博C107

题目：Efficient Approaches for Two Big Data Matrix Optimization Problems
报告人：吕召松教授（Simon Fraser University, Canada）
地点：屏峰校区博C107
时间：2015年6月10日（周三）下午3:30-4:30
摘要：In the first part of this talk, we consider low rank matrix completion problem, which has wide applications such as collaborative filtering, image inpainting and Microarray data imputation. We present an efficient and scalable algorithm for matrix completion. In each iteration, we pursue a rank-one matrix basis generated by the top singular vector pair of the current approximation residual and update the weights for all rank-one matrices obtained up to the current iteration. We further propose a novel weight updating rule to reduce the time and storage complexity, making the proposed algorithm scalable to large matrices. We establish a linear rate of convergence for the algorithm. Numerical experiments demonstrate that our algorithm is much more efficient than the state-of-the-art algorithms while achieving similar or better prediction performance.
 
In the second part we consider the problem of estimating multiple graphical models simultaneously using the fused lasso penalty, which encourages adjacent graphs to share similar structures. One important application of this problem is for the analysis of brain networks of Alzheimer's disease. We establish a necessary and sufficient condition for the graphs to be decomposable. As a consequence, a simple but effective screening rule is proposed, which decomposes large graphs into small subgraphs and dramatically reduces the overall computational cost. Numerical experiments demonstrate the effectiveness and efficiency of our proposed approach.
 
报告人简介：吕召松，加拿大西蒙弗雷泽大学终身教授。2005年获美国佐治亚理工学院运筹学博士学位。主要研究兴趣在大规模连续优化问题的理论和算法，及其在数据挖掘、信号处理等领域的应用。目前担任SIAM Journal on Optimization, Big Data and Information Analytics杂志副主编，主持加拿大自然科学与工程技术研究理事会（NSERC）基金3项，发表学术论文40余篇，其中20余篇发表在运筹与优化国际顶级期刊Mathematical Programming和SIAM Journal on Optimization等上。