一种基于局部保持的隐变量模型  

A Latent Variable Model Based on Local Preservation

作　　者：王秀美[1,2] 高新波[2] 张乾坤[2] 宋国乡[1] 

机构地区：[1]西安电子科技大学理学院,西安710071 [2]西安电子科技大学电子工程学院,西安710071

出　　处：《模式识别与人工智能》2010年第3期369-375,共7页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金项目(No.60702061;6077106;60702061);教育部长江学者和创新团队支持计划项目(No.IRT0645)资助

摘　　要：隐变量模型是一类有效的降维方法,但是由非线性核映射建立的隐变量模型不能保持数据空间的局部结构.为了克服这个缺点,文中提出一种保持数据局部结构的隐变量模型.该算法充分利用局部保持映射的保局性质,将局部保持映射的目标函数作为低维空间中数据的先验信息,对高斯过程隐变量中的低维数据进行约束,建立局部保持的隐变量.实验结果表明,相比原有的高斯过程隐变量,文中算法较好地保持数据局部结构的效果.Latent variable model （LVM） is a kind of efficient nonlinear dimensionality reduction algorithm through establishing smooth kernel mappings from the latent space to the data space. However, this kind of mappings cannot keep the points close in the latent space even they are close in data space. A LVM is proposed based on locality preserving projection （LPP） which can preserve the locality structure of dataset. The objective function of LPP is considered as a prior of the variables in the Gaussian process latent variable model （GP-LVM）. The proposed locality preserving GP-LVM is built with the constrained term of the objective function. Compared with the traditional LPP and GP-LVM, experimental results show that the proposed method performs better in preserving local structure on common data sets.

关 键 词：降维 隐变量模型(LVM) 局部距离保持 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]