检索结果-南通市图书馆

A convergence algorithm for graph co-regularized transfer learning

Science China(Information Sciences) 2023年第3期66卷 150-159页

作者： Zuyuan YANG naiyao liang Zhenni LI Shengli XIE Guangdong Key Laboratory of Io T Information Technology School of Automation Guangdong University of Technology Key Laboratory of iDetection and Manufacturing-IoT Ministry of Education Guangdong-Hong Kong-Macao Joint Laboratory for Smart Discrete Manufacturing

Transfer learning is an important technology in addressing the problem that labeled data in a target domain are difficult to collect using extensive labeled data from the source domain. Recently,an algorithm named graph co-regularized transfer learning(GTL) has shown a competitive performance in transfer learning. However, its convergence is affected by the used approximate scheme, degenerating learned results. In this paper, after analyzing convergence conditions, we propose a novel update rule using the multiplicative update rule and develop a new algorithm named improved GTL(IGTL) with a strict convergence guarantee. Moreover, to prove the convergence of our method, we design a special auxiliary function whose value is intimately related to that of the objective function. Finally, the experimental results on the synthetic dataset and two real-world datasets confirm that the proposed IGTL is convergent and performs better than the compared methods.

关键词： transfer learning convergence analysis non-negative matrix factorization multiplicative update algorithm optimization

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Label-noise robust classification with multi-view learning

引用

Science China(Technological Sciences) 2023年第6期66卷 1841-1854页

作者： liang naiyao YANG ZuYuan LI LingJiang LI ZhenNi XIE ShengLi Guangdong Key Laboratory of IoT Information Technology School of AutomationGuangdong University of TechnologyGuangzhou 510006China Key Laboratory of iDetection and Manufacturing-IoT Ministry of EducationGuangzhou 510006China Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou 510006China

Label noise is often contained in the training data due to various human factors or measurement errors,which significantly causes a negative effect on *** many previous methods that have been proposed to learn robust classifiers,they are mainly based on the single-view *** the other hand,although existing multi-view classification methods benefit from the more comprehensive information,they rarely consider label *** this paper,we propose a novel label-noise robust classification model with multi-view learning to overcome these *** the proposed model,not only the classifier learning but also the label-noise removal can benefit from the multi-view ***,we relax the label matrix of the basic multi-view least squares regression model,and develop a nonlinear transformation with a natural probabilistic approximation in the process of labels,which is conveniently optimized and beneficial to improve the discriminative ability of ***,we preserve the intrinsic manifold structure of multi-view data on the relaxed label matrix,facilitating the process of label *** optimizing the proposed model with the nonlinear transformation,we derive a lemma about the partial derivation of the softmax related function,and develop an efficient alternating *** evaluations on six real-world datasets confirm the advantages of the proposed method,compared to the related state-of-the-art methods.

关键词： label noise multi-view learning classification robust least squares regression label relaxation

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Diverse Deep Matrix Factorization With Hypergraph Regularization for Multi-View Data Representation

引用

自动化学报:英文版 2023年第11期10卷 2154-2167页

作者： Haonan Huang Guoxu Zhou naiyao liang Qibin Zhao Shengli Xie School of Automation Guangdong University of Technologyand also with the Key Laboratory of Intelligent Detection and the Internet of Things in ManufacturingMinistry of EducationGuangzhou 510006China School of Automation Guangdong University of TechnologyGuangzhou 510006China RIKEN AIP Japanand also with the School of AutomationGuangdong University of TechnologyGuangzhou 510006China 111 Center for Intelligent Batch Manufacturing Based on IoT Technology(GDUT) and also with the Guangdong-Hong Kong-Macao Joint Laboratory for Smart Discrete ManufacturingGuangdong University of TechnologyGuangzhou 510006China IEEE

Deep matrix factorization(DMF)has been demonstrated to be a powerful tool to take in the complex hierarchical information of multi-view data(MDR).However,existing multiview DMF methods mainly explore the consistency of multi-view data,while neglecting the diversity among different views as well as the high-order relationships of data,resulting in the loss of valuable complementary *** this paper,we design a hypergraph regularized diverse deep matrix factorization(HDDMF)model for multi-view data representation,to jointly utilize multi-view diversity and a high-order manifold in a multilayer factorization framework.A novel diversity enhancement term is designed to exploit the structural complementarity between different views of *** regularization is utilized to preserve the high-order geometry structure of data in each *** efficient iterative optimization algorithm is developed to solve the proposed model with theoretical convergence *** results on five real-world data sets demonstrate that the proposed method significantly outperforms stateof-the-art multi-view learning approaches.

关键词： Deep matrix factorization(DMF) diversity hypergraph regularization multi-view data representation(MDR)

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：