高维数据空间的协方差矩阵迹的比率相合性

在处理高维数据的检验和分类等问题时,涉及到协方差矩阵的估计.而在高维数据领域,协方差矩阵估计的精度将对诸如检验和分类等问题起到非常重要的影响.主要考虑多样本条件下协方差矩阵的比率相合性问题,证明了两样本和三样本情况下的高维数据协方差矩阵比率相合性.     

基于非线性变换的PCP-C模型及其在地下水动态分类中的应用

为了克服目前地下水动态分类方法中存在的不能揭示分类指标空间到类型空间的非线性映射关系、方法复杂、计算量大等缺陷,可采用基于非线性变换的主成分投影(PCP)-聚类(C)模型,对地下水动态进行分类.方法首先对分类指标数据进行对数中心化变换,然后应用主成分投影法将变换后的多维指标向量映射到最优一维向量空间,并根据各样本指标在一维向量空间的投影值进行聚类分析,由此得到地下水动态分类结果.地下水动态分类结果表明,建议方法概念清晰,结构简单,计算简便,分类结果可信,是一种有效的地下水动态分类方法.     

基于数据流形结构的聚类方法及其应用研究

随着信息社会的不断发展,人类已经进入了信息爆炸时代,海量的数据使数据处理变得繁琐复杂,因此如何对现有的高维数据降维、聚类,并在一定程度上消除高维数据中存在的噪声是解决该问题的关键.基于相关的理论知识采用先降维后聚类的步骤,把高维数据按照子空间结构和流形结构两种情况分类,运用稀疏子空间聚类、谱多流形聚类、K-manifolds方法进行建模求解,通过对各种方法的对比,得出谱多流形聚类方法运行速度快,聚类准确度高,是最具有一般性特征的模型.     

矩阵奇异值分解理论在中文文本分类中的应用

用改进的截断与转换的矩阵奇异值分解算法,设计实现了基于字频特征的中文文本分类器.理论分析与实验结果表明,采用的方法提高了数值计算精度,降低了文本集特征空间的维数,简化了文本分类算法的时间复杂度,提高了文本分类准确率.     

基于自编码器的卷积神经网络肿瘤亚型分类研究

根据临床收录的肿瘤基因表达谱数据,可以利用分类器进行肿瘤亚型分类.由于基因表达谱数据样本小、维度高,难以提取有效特征,分类效果往往不好,而且很容易过拟合.针对这些问题,研究利用自编码器对特征基因进行降维,并构建多尺度的神经网络进行学习分类,综合考虑不同尺度的特征,提出A-CNNs网络,不仅解决了高维样本难以处理的问题,且有效避免了纵向加深神经网络带来的过拟合,得到了较高的平均分类精度,并与其他机器学习方法进行对比实验,实验证明所构建的分类模型可以取得较佳的分类效果.     

手部软组织红外热像图处理与K-L变换分析

首先对手部软组织的红外热像图进行图像处理,包括灰度化、降噪、增强、分割,其中图像分割采用Otsu算法选择适当的阈值,从而将目标图像与背景分割.然后将处理后图像的像素值代入K-L变换,得到样本的特征空间,将所有样本都投影到该特征空间得到投影系数.最后利用欧式距离公式,最终判断手部软组织的病变情况.     

投影寻踪动态聚类模型在房地产投资环境评价中的应用

将投影寻踪动态聚类模型引入到房地产投资环境评价方法中.针对房地产投资环境评价所面临的多因素高维复杂性问题,该模型能够完全根据样本数据特性将高维数据通过投影向量投影到低维数据,同时实现对低维数据的排序和自动聚类分析,进而通过研究低维数据以实现对高维数据的研究.最后通过辽宁省工业地产投资环境评价实例验证了该模型在房地产投资环境评价中的适用性.     

基于SVM的客车车型分类

通过支持向量机(SVM)对客车车型的长,宽,高,宽长比等7个特征进行特征选择,得到的准确率最高的子集是长、宽、高、宽长比、宽高比,以它作为样本特征进行分类.对客车的4类车型进行分类,每类车型选择80个样本,50个样本进行训练,30个样本进行预测,结果表明:对1类车型的分类准确率可达到100%,对2类和4类车型可达到96%以上,对3类车可达到93%以上.得到了比选用长、宽、高作为特征进行分类更优的结果.然后运用加入参数寻优的SVM对客车的4类车型进行分类,并加以比较.基于高斯函数的特性,两次用到SVM进行机器学习时,核函数均选用RBF核函数.     

一种基于正交函数系的时间序列聚类方法

基于正交函数系和FCM算法,提出了一种新的时间序列聚类的方法.该方法首先通过一个非线性映射,将长度为n的时间序列映射到L_2空间,然后通过计算函数之间的距离得到时间序列之间的相似度.在此基础上,经过FCM算法实现时间序列的聚类.该方法克服了时间序列的高维数特征为时间序列聚类带来的计算困难.实验结果表明,对高维的时间序列,该方法在压缩率达到80%的情况下,依然具有良好的聚类效果.     

空间计量模型中权重矩阵的类型与选择

根据空间效应产生起点及理论基础的不同,归纳了现有空间计量文献中邻接矩阵、反距离矩阵、经济特征矩阵以及嵌套矩阵等主要权重形式,并总结了其共同点、优缺点、演变脉络及使用注意事项.针对截面式权重矩阵本身面临的限制构造了两种必要的转换,即通过转换实现对不同地理区域之间空间效应的考察,以及从截面权重到面板权重的转换.最后指出研究者应该尽量采用多种新方法来确定空间权重形式以使其更客观.     

Special bases for solutions of a generalized Maxwell system in 3‐dimensional space

The aim of this paper is to study complete polynomial systems in the kernel space of conformally invariant differential operators in higher spin theory. We investigate the kernel space of a generalized Maxwell operator in 3‐dimensional space. With the already known decomposition of its homogeneous kernel space into 2 subspaces, we investigate first projections from the homogeneous kernel space to each subspace. Then, we provide complete polynomial systems depending on the given inner product for each subspace in the decomposition. More specifically, the complete polynomial system for the homogenous kernel space is an orthogonal system wrt a given Fischer inner product. In the case of the standard inner product in L² on the unit ball, the provided complete polynomial system for the homogeneous kernel space is a partially orthogonal system. Further, if the degree of homogeneity for the respective subspaces in the decomposed kernel spaces approaches infinity, then the angle between the 2 subspaces approaches π/2.     

A matrix modular neural network based on task decomposition with subspace division by adaptive affinity propagation clustering

In this paper, a matrix modular neural network (MMNN) based on task decomposition with subspace division by adaptive affinity propagation clustering is developed to solve classification tasks. First, we propose an adaptive version to affinity propagation clustering, which is adopted to divide each class subspace into several clusters. By these divisions of class spaces, a classification problem can be decomposed into many binary classification subtasks between cluster pairs, which are much easier than the classification task in the original multi-class space. Each of these binary classification subtasks is solved by a neural network designed by a dynamic process. Then all designed network modules form a network matrix structure, which produces a matrix of outputs that will be fed to an integration machine so that a classification decision can be made. Finally, the experimental results show that our proposed MMNN system has more powerful generalization capability than the classifiers of single 3-layered perceptron and modular neural networks adopting other task decomposition techniques, and has a less training time consumption.     

Energy discriminant analysis, quantum logic, and fuzzy sets

In this paper, we show that quantum logic of linear subspaces can be used for recognition of random signals by a Bayesian energy discriminant classifier. The energy distribution on linear subspaces is described by the correlation matrix of the probability distribution. We show that the correlation matrix corresponds to von Neumann density matrix in quantum theory. We suggest the interpretation of quantum logic as a fuzzy logic of fuzzy sets. The use of quantum logic for recognition is based on the fact that the probability distribution of each class lies approximately in a lower-dimensional subspace of feature space. We offer the interpretation of discriminant functions as membership functions of fuzzy sets. Also, we offer the quality functional for optimal choice of discriminant functions for recognition from some class of discriminant functions.     

酉空间中一类子空间的排列问题与具有纠错能力的d~z-析取矩阵

Pooling设计在实践中有着广泛的应用,它的数学模型是d~z-析取矩阵.本文利用酉空间的一类子空间构做了一类新的d~z-析取矩阵.为了讨论此设计的纠错能力,重点研究了酉空间中的一类子空间的排列问题,即对于酉空间F_q~2~((n))上的一个给定的(m,s)型子空间C和一个整数d,找到C的d个(m-1,s-1)型子空间H_1,H_2,…,H_d,使得包含在H_1∪H_2∪…∪H_d中的(r,s-4)型子空间的个数最多,并确定这个数的上界.然后应用此结果,给出了d~z-析取矩阵中反映纠错能力的z值的紧界.     

Krylov eigenvalue strategy using the FEAST algorithm with inexact system solves

The FEAST eigenvalue algorithm is a subspace iteration algorithm that uses contour integration to obtain the eigenvectors of a matrix for the eigenvalues that are located in any user‐defined region in the complex plane. By computing small numbers of eigenvalues in specific regions of the complex plane, FEAST is able to naturally parallelize the solution of eigenvalue problems by solving for multiple eigenpairs simultaneously. The traditional FEAST algorithm is implemented by directly solving collections of shifted linear systems of equations; in this paper, we describe a variation of the FEAST algorithm that uses iterative Krylov subspace algorithms for solving the shifted linear systems inexactly. We show that this iterative FEAST algorithm (which we call IFEAST) is mathematically equivalent to a block Krylov subspace method for solving eigenvalue problems. By using Krylov subspaces indirectly through solving shifted linear systems, rather than directly using them in projecting the eigenvalue problem, it becomes possible to use IFEAST to solve eigenvalue problems using very large dimension Krylov subspaces without ever having to store a basis for those subspaces. IFEAST thus combines the flexibility and power of Krylov methods, requiring only matrix–vector multiplication for solving eigenvalue problems, with the natural parallelism of the traditional FEAST algorithm. We discuss the relationship between IFEAST and more traditional Krylov methods and provide numerical examples illustrating its behavior.     

Factoring matrices into the product of two matrices

Linear algebra of factoring a matrix into the product of two matrices with special properties is developed. This is accomplished in terms of the so-called inverse of a matrix subspace which yields an extended notion for the invertibility of a matrix. The product of two matrix subspaces gives rise to a natural generalization of the concept of matrix subspace. Extensions of these ideas are outlined. Several examples on factoring are presented. AMS subject classification (2000)  15A23, 65F30     

Inner-approximating domains of attraction for discrete-time switched systems via Multi-step multiple Lyapunov-like functions

In this paper, we propose an iterative approach for estimating the domains of attraction for a class of discrete-time switched systems, where the state space is divided into several disjoint regions and each region is described by polynomial inequalities. At first, we introduce the basic concepts of Multi-step state subsequence, Multi-step state subspace, Multi-step basin of attraction and Multi-step multiple Lyapunov-like function. Secondly, beginning with an initial inner estimation, a theoretical framework is proposed for estimating the domain of attraction by iteratively calculating the Multi-step multiple Lyapunov-like functions. Thirdly, notice that the Multi-step state subspaces may be empty sets such that the corresponding constraints in the theoretical framework are redundant, we propose a numerical approach based on the homotopy continuation method to pre-check the non-emptiness of the Multi-step state subspaces, and then under-approximatively realize the framework by using S-procedure and sum of squares programming. At last, we implement our iterative approach and apply it to three discrete-time switched system examples with comparisons to existing methods in the literatures. These computation and comparison results show the advantages of our method.     

Perturbation analysis of singular subspaces and deflating subspaces

Summary. Perturbation expansions for singular subspaces of a matrix and for deflating subspaces of a regular matrix pair are derived by using a technique previously described by the author. The perturbation expansions are then used to derive Fr\'echet derivatives, condition numbers, and th-order perturbation bounds for the subspaces. Vaccaro's result on second-order perturbation expansions for a special class of singular subspaces can be obtained from a general result of this paper. Besides, new perturbation bounds for singular subspaces and deflating subspaces are derived by applying a general theorem on solution of a system of nonlinear equations. The results of this paper reveal an important fact: Each singular subspace and each deflating subspace have individual perturbation bounds and individual condition numbers. Received July 26, 1994     

Subspace choices for the Celis-Dennis-Tapia problem

Grapiglia et al. (2013) proved subspace properties for the Celis-Dennis-Tapia (CDT) problem. If a subspace with lower dimension is appropriately chosen to satisfy subspace properties, then one can solve the CDT problem in that subspace so that the computational cost can be reduced. We show how to find subspaces that satisfy subspace properties for the CDT problem, by using the eigendecomposition of the Hessian matrix of the objection function. The dimensions of the subspaces are investigated. We also apply the subspace technologies to the trust region subproblem and the quadratic optimization with two quadratic constraints.     

On the nonnegative self-adjoint solutions of the operator Riccati equation for infinite dimensional systems

The nonnegative self-adjoint solutions of the operator Riccati equation (ORE) are studied for stabilizable semigroup Hilbert state space systems with bounded sensing and control. Basic properties of the maximal solution of the ORE are investigated: stability of the corresponding closed loop system, structure of the kernel, Hilbert-Schmidt property. Similar properties are obtained for the nonnegative self-adjoint solutions of the ORE. The analysis leads to a complete classification of all nonnegative self-adjoint solutions, which is based on a bijection between these solutions and finite dimensional semigroup invariant subspaces contained in the antistable unobservable subspace.