多维标度问题的一类直接拟合法
A CLASS OF DIRECT FITTING METHOD FOR MULTIDIMENSIONAL SCALING PROBLEM
查看参考文献15篇
|
文摘
|
多维标度分析(MDS)是一种用于分析和可视化数据之间相似性或距离关系的统计方法.它通过将数据点映射到低维空间中的坐标来表示它们之间的相对距离或相似性.多维标度问题的古典解通过对(非欧氏型)距离矩阵平方进行双中心化处理,进而通过截断特征值分解寻求低维的拟合构造点.本文对距离矩阵平方进行直接拟合.重构问题为零列和Stiefel子流形和线性流形约束下的矩阵优化模型,并结合乘积流形几何性质,设计一类自适应问题模型的基于Zhang-Hager技术拓展的黎曼梯度下降求解算法.数值实验说明通过直接拟合能得到误差更小的拟合欧氏距离矩阵,且所提算法与已有投影梯度流算法及黎曼优化工具箱中的黎曼一阶和二阶算法在迭代效率上有一定的优势. |
|
其他语种文摘
|
Multidimensional scaling (MDS) is a statistical method used to analyze and visualize the similarity or distance relationships between data points. It represents the relative distances or similarities between data points by mapping them to coordinates in a low-dimensional space. The classical solution to the multidimensional scaling problem involves a double centering process on the squared (non-Euclidean) distance matrix, followed by truncating the eigenvalue decomposition to seek a low-dimensional approximate configuration of points. In this paper, we directly fit the squared distance matrix and reformulate the reconstruction problem as a constrained matrix optimization model in the product manifold composed of zero column and column orthogonal matrices and diagonal matrices. By leveraging the geometric properties cf the product manifold and incorporating an extended Riemannian gradient descent algorithm based on Zhang-Hager technique, we design a class of adaptive problem models. Numerical experiments demonstrate that direct fitting yields a smaller error in fitting the Euclidean distance matrix. Moreover, the proposed algorithm exhibits certain advantages in terms of iterative efficiency compared to existing projection gradient flow algorithms and first-order and second-order Riemannian algorithms in the Riemannian optimization toolbox. |
|
来源
|
计算数学
,2024,46(3):291-311 【核心库】
|
|
DOI
|
10.12286/jssx.j2023-1132
|
|
关键词
|
多维标度分析
;
直接拟合
;
黎曼梯度法
;
乘积流形
|
|
地址
|
1.
桂林电子科技大学数学与计算科学学院, 桂林, 541004
2.
云南大学数学与统计学院, 昆明, 650000
3.
桂林电子科技大学数学与计算科学学院,广西应用数学中心(桂林电子科技大学), 广西高校数据分析与计算重点实验室, 桂林, 541004
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
0254-7791 |
|
学科
|
数学 |
|
基金
|
国家自然科学基金
;
广西自然科学基金
;
2023年广西研究生教肓创新计划项目
;
2022年桂林电子科技大学位与研究生教育改革项目
;
广西自动检测技术与仪器重点实验室基金
;
广西科技项目
|
|
文献收藏号
|
CSCD:7809925
|
参考文献 共
15
共1页
|
|
1.
Browne M. The Young-Householder algorithm and the least squares multidimensional scaling of squared distances.
Journal of Classification,1987,4:175-190
|
CSCD被引
1
次
|
|
|
|
|
2.
Trendafilov N T. DINDSCAL: Direct INDSCAL.
Statistics and Computing,2012,22:445-454
|
CSCD被引
1
次
|
|
|
|
|
3.
Trendafilov N.
Multivariate Data Analysis on Matrix Manifolds (with Manopt),2021
|
CSCD被引
1
次
|
|
|
|
|
4.
Absil P A.
Optimization algorithms on matrix manifolds,2008
|
CSCD被引
35
次
|
|
|
|
|
5.
Absil P A. An extrinsic look at the Riemannian Hessian.
International conference on geometric science of information,2013:361-368
|
CSCD被引
5
次
|
|
|
|
|
6.
Nocedal J.
Numerical optimization,1999
|
CSCD被引
54
次
|
|
|
|
|
7.
Grippo L. A nonmonotone line search technique for Newton's method.
SIAM Journal on Numerical Analysis,1986,23(4):707-716
|
CSCD被引
122
次
|
|
|
|
|
8.
Zhang H. A nonmonotone line search technique and its application to unconstrained optimization.
SIAM Journal on Optimization,2004,14(4):1043-1056
|
CSCD被引
82
次
|
|
|
|
|
9.
Dai Y H. Conjugate Gradient Methods with Armijo-type Line Searches.
Acta Mathematicae Applicatae Sinica,2002,18(1):123-130
|
CSCD被引
12
次
|
|
|
|
|
10.
Wen Z. A feasible method for optimization with orthogonality constraints.
Mathematical Programming,2013,142(1/2):397-434
|
CSCD被引
43
次
|
|
|
|
|
11.
Boumal N. Manopt, a Matlab toolbox for optimization on manifolds.
The Journal of Machine Learning Research,2014,15(1):1455-1459
|
CSCD被引
12
次
|
|
|
|
|
12.
Oviedo H. Global convergence of Riemannian line search methods with a Zhang-Hager-type condition.
Numerical Algorithms,2022,91(3):1183-1203
|
CSCD被引
4
次
|
|
|
|
|
13.
Li J F. A Riemannian optimization approach for solving the generalized eigenvalue problem for nonsquare matrix pencils.
Journal of Scientific Computing,2020,82:1-43
|
CSCD被引
8
次
|
|
|
|
|
14.
Zhu X. A Riemannian conjugate gradient method for optimization on the Stiefel manifold.
Computational optimization and Applications,2017,67:73-110
|
CSCD被引
14
次
|
|
|
|
|
15.
Dai Y H. A nonmonotone conjugate gradient algorithm for unconstrained optimization.
Journal of System Scinence and Comeplexity,2002,15(2):139
|
CSCD被引
1
次
|
|
|
|
|
了解气象卫星更多信息，请访问