Geometric Distance Between Positive Definite Matrices of Different Dimensions

被引:13
|
作者
Lim, Lek-Heng [1 ]
Sepulchre, Rodolphe [2 ]
Ye, Ke [3 ]
机构
[1] Univ Chicago, Dept Stat, Computat & Appl Math Initiat, Chicago, IL 60637 USA
[2] Univ Cambridge, Dept Engn, Cambridge CB2 1PZ, England
[3] Chinese Acad Sci, Acad Math & Syst Sci, KLMM, Beijing 100190, Peoples R China
基金
美国国家科学基金会; 国家重点研发计划; 中国国家自然科学基金; 欧洲研究理事会;
关键词
Riemannian manifold; geodesic distance; positive definite matrices; covariance matrices; ellipsoids; RIEMANNIAN GEOMETRY;
D O I
10.1109/TIT.2019.2913874
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We show how the geodesic distance on S-++(n), the cone of n x n real symmetric or complex Hermitian positive definite matrices regarded as a Riemannian manifold, may be used to naturally define a distance between two such matrices of different dimensions. Given that S-++(n) also parameterizes n-dimensional ellipsoids, inner products on R-n, and n x n covariances of nondegenerate probability distributions, this gives us a natural way to define a geometric distance between a pair of such objects of different dimensions.
引用
收藏
页码:5401 / 5405
页数:5
相关论文
共 50 条