文献类型：期刊文献

中文题名：基于类间和类内密度的多视角距离度量学习

英文题名：Multi-view Distance Metric Learning with Inter-class and Intra-class Density

作者：任双艳[1];郭威[1];范昌琪[2];王喆[1];吴松洋[3]

机构：[1]华东理工大学信息科学与工程学院,上海200237;[2]上海移动互联网产业促进中心,上海200333;[3]公安部第三研究所,上海201204

年份：2022

卷号：49

期号：S02

起止页码：77

中文期刊名：计算机科学

外文期刊名：Computer Science

收录：CSTPCD;;北大核心:【北大核心2020】；CSCD:【CSCD_E2021_2022】；

基金：上海市科技计划项目(20511100600);国家自然科学基金(62076094);信息网络安全公安部重点实验室开放课题项目(C20603)

语种：中文

中文关键词：几何信息;类间密度;类内密度;互补信息;视角相关性

外文关键词：Geometric information;Inter-class density;Intra-class density;Complementary information;View correlation

摘要：几何信息可以为分类方法提供先验知识和直观解释。从几何角度观察样本是一种新的样本学习方法,密度则是几何信息中非常直观的表现形式。提出了基于类间和类内密度的多视角距离度量学习方法来学习一个度量空间。在这个空间内,异类样本更加分散,同类样本更加紧密。首先,在大边际框架下引入类间密度,通过最小化类间密度来约束度量空间中的样本,从而实现类间分散,提高分类性能。其次,引入类内密度,通过最大化类内密度来达到同类样本互相靠近的效果,从而实现类内紧凑。最后,为了更好地挖掘多视角样本的互补信息,最大化度量空间中各视角之间的相关性,使各视角自适应地相互学习,探索视角之间的互补信息。在真实数据集上的大量实验结果证明了该方法的优越性。
Geometric information can provide prior knowledge and intuitive explanation for classification methods.Observing samples from geometric perspective is a novel method of sample learning,and density is a very intuitive form of geometric information.This paper proposes a multi-view distance metric learning method with inter-class and intra-class density to learn a metric space.In this space,the heterogeneous samples are more scattered,and the homogeneous samples are closer.First,the inter-class density is introduced under the large margin framework,and the samples in the metric space are constrained by minimizing the inter-class density,so as to realize the inter-class dispersion and improve the classification performance.Second,maximize the intraclass density to achieve the effect of similar samples close to each other,so as to achieve intra-class compactness.Finally,to better mine the complementary information of the multi-view samples,the correlation between the views in the metric space is maximized,so that the views can learn from each other adaptively and explore the complementary information among the views.A large number of experimental results on real-world datasets demonstrate the superiority of the proposed method.