[1]江帆,田青.基于结构化稀疏投影的多视图特征提取框架[J].计算机技术与发展,2019,29(03):6-11.[doi:10.3969/ j. issn.1673-629X.2019.03.002]
　JIANG Fan,TIAN Qing.A Framework of Multi-view Feature Extraction Based on Multiple Structured Sparsity Projection[J].,2019,29(03):6-11.[doi:10.3969/ j. issn.1673-629X.2019.03.002]
点击复制

基于结构化稀疏投影的多视图特征提取框架()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年03期
页码:
6-11
栏目:
智能、算法、系统工程
出版日期:
2019-03-10

文章信息/Info

Title:
A Framework of Multi-view Feature Extraction Based on Multiple Structured Sparsity Projection
文章编号:
1673-629X(2019)03-0006-06
作者:
江帆1 ; 田青2
1. 南京航空航天大学 计算机科学与技术学院,江苏 南京 211106;2. 南京信息工程大学 计算机与软件学院,江苏 南京 210044
Author(s):
JIANG Fan1 ; TIAN Qing2
1. School of Computer Science and Technology,Nanjing University of Aeronautics &Astronautics, Nanjing 211106,China;2. School of Computer and Software,Nanjing University of Information Science and Technology, Nanjing 210044,China
关键词:
一致性; 互补性; 多视图学习; 特征提取; 结构稀疏; Stiefel 流形
Keywords:
consistency; complementarity; multi-view learning; feature extraction; structured sparsity; Stiefel manifold
分类号:
TP391
DOI:
10.3969/ j. issn.1673-629X.2019.03.002
摘要:
多视图学习通常基于两个重要原则:一致性原则和互补性原则。 视图的一致性源自所有视图间的共享信息;而视图的互补性源自不同视图的特有信息。 然而,现有多视图学习算法通常只关注其中的一种,以致得到的模型性能非最优。为此,提出一种新的多视图特征提取算法框架,称为多重结构化稀疏投影(MSSP),其能同时提取视图的公共和特有信息。MSSP 的目标函数包含两个部分:融合投影的判别项和联合投影矩阵的结构化稀疏正则化项。 通过以上建模方式,不同视图的一致和互补信息同时得到了融合利用。 此外,给出所建模型基于在 Stiefel 流形上的梯度下降的求解算法,获得局部最优解,并分析推测了模型参数随投影维数的变化趋势。 相关对比实验验证了 MSSP 的建模有效性和参数的变化趋势。
Abstract:
Multi-view learning is usually based on two important principles:consistency and complementarity. Consistency derives from the shared information between views and complementarity derives from the unique information of different views. However,the existing multi-view learning algorithms usually focus on only one of them,so that the resulting model performance is not optimal. For this,we propose a new multi-view feature extraction algorithm framework called multiple structured sparsity projection (MSSP),which can sim ultaneously extract the common and unique information of the view. The objective function of MSSP contains two parts:fusion projection discriminant and structured sparsity regularization. Then the consistent and complementary information of different views can be fused simultaneously. In addition,we also give an algorithm based on the gradient decent over the Stiefel manifold to obtain a local optimal so lution,and analyze how the hyper-parameter change with the dimension of the projection. The experiment verifies the effectiveness of MSSP and the trend of the hyper-parameter.

相似文献/References:

[1]李玲娟 豆坤.层次分析法中判断矩阵的一致性研究[J].计算机技术与发展,2009,(10):131.
　LI Ling-juan,DOU Kun.Research on the Consistency of the Judgment Matrix in AHP[J].,2009,(03):131.
[2]王金环 李宝敏.基于本体DL的语义推理研究[J].计算机技术与发展,2009,(11):94.
　WANG Jin-huan,LI Bao-min.Study of Semantic Reasoning Based on Ontology Description Logic[J].,2009,(03):94.
[3]陈堃 李心科.基于一致性度量的属性约简的研究[J].计算机技术与发展,2008,(10):64.
　CHEN Kun,LI Xin-ke.Research of Attribute Reduction Based on Consistency Measure[J].,2008,(03):64.
[4]彭俊 谢荣传 王大刚 耿波.一种基于相似性的规则集一致性度量的新方法[J].计算机技术与发展,2008,(11):73.
　PENG Jun,XIE Rong-chuan,WANG Da-gang,et al.A New Measuring Rule Set Consistency Method Based on Similarity[J].,2008,(03):73.
[5]李彬 任国林.Linux内核基于对称多处理机的实现分析[J].计算机技术与发展,2006,(01):129.
　LI Bin,PEN Guo-lin.Analysis of Linux Kernel Based on SMP[J].,2006,(03):129.
[6]司慧勇.一种用于科研协同的电子白板系统的设计与实现[J].计算机技术与发展,2006,(10):219.
　SI Hui-yong.Design and Implementation of White Board in Scientific Research Collaboration Environment[J].,2006,(03):219.
[7]张灵峰 夏战锋 彭志平.基于Tbox和Abox的描述逻辑推理研究[J].计算机技术与发展,2010,(11):122.
　ZHANG Ling-feng,XIA Zhan-feng,PENG Zhi-ping.Research on Description Logic＇s Reasoning Based on Tbox and Abox[J].,2010,(03):122.
[8]曹黎侠 冯孝周.新的改进AHP算法研究及应用[J].计算机技术与发展,2010,(12):115.
　CAO Li-xia,FENG Xiao-zhou.A New Improved AHP Algorithm Research and Application[J].,2010,(03):115.
[9]肖庆 焦健.网络安全策略与措施的语义一致性研究[J].计算机技术与发展,2011,(08):220.
　XIAO Qing,JIAO Jian.Research on Semantic Consistency of Network Security Policy and Measure[J].,2011,(03):220.
[10]陈荟慧 王伟静.基于UML的数据流图可视化编辑工具的设计[J].计算机技术与发展,2012,(05):145.
　CHEN Hui-hui,WANG Wei-jing.Design of Visual Edit Tool for DFD Based on UML[J].,2012,(03):145.

更新日期/Last Update: 2019-03-10