Geometric foundations for scaling-rotation statistics on symmetric positive definite matrices: minimal smooth scaling-rotation curves in low dimensions

DOI10.1214/17-EJS1250MaRDI QIDQ521342zbMATH OpenOpenAlexFDO

Authors David Groisser, Sungkyu Jung, Armin Schwartzman

Publication date 7 April 2017

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1602.01187, https://projecteuclid.org/euclid.ejs/1491292817

zbMATH Keywords

eigen-decompositiongeodesicsscaling-rotation distancestatistics on manifoldsstratified spacessymmetric group

Mathematics Subject Classification ID

Eigenvalues, singular values, and eigenvectors (15A18) Orthogonal and unitary groups in metric geometry (51F25) General geometric structures on manifolds (almost complex, almost product structures, etc.) (53C15) Geodesics in global differential geometry (53C22) Global differential geometry (53C99)

Abstract: We investigate a geometric computational framework, called the "scaling-rotation framework", on ${mSym}^{+}(p)$, the set of $pimesp$ symmetric positive-definite (SPD) matrices. The purpose of our study is to lay geometric foundations for statistical analysis of SPD matrices, in situations in which eigenstructure is of fundamental importance, for example diffusion-tensor imaging (DTI). Eigen-decomposition, upon which the scaling-rotation framework is based, determines both a stratification of ${mSym}^{+}(p)$, defined by eigenvalue multiplicities, and fibers of the "eigen-composition" map $SO(p)imes{mDiag}^{+}(p)o{mSym}^{+}(p)$. This leads to the notion of scaling-rotation distance [Jung et al. (2015)], a measure of the minimal amount of scaling and rotation needed to transform an SPD matrix, $X,$ into another, $Y,$ by a smooth curve in ${mSym}^{+}(p)$. Our main goal in this paper is the systematic characterization and analysis of minimal smooth scaling-rotation (MSSR) curves, images in ${mSym}^{+}(p)$ of minimal-length geodesics connecting two fibers in the "upstairs" space $SO(p)imes{mDiag}^{+}(p)$. The length of such a geodesic connecting the fibers over $X$ and $Y$ is what we define to be the scaling-rotation distance from $X$ to $Y.$ For the important low-dimensional case $p=3$ (the home of DTI), we find new explicit formulas for MSSR curves and for the scaling-rotation distance, and identify $calM(X,Y)$ in all "nontrivial" cases. The quaternionic representation of $SO(3)$ is used in these computations. We also provide closed-form expressions for scaling-rotation distance and MSSR curves for the case $p=2$.

Recommendations

• Scaling-rotation distance and interpolation of symmetric positive-definite matrices
• Uniqueness questions in a scaling-rotation geometry on the space of symmetric positive-definite matrices
• SPD tensors regularization via Iwasawa decomposition
• Coordinates-Based Diffusion Over the Space of Symmetric Positive-Definite Matrices
• Riemannian geometry for the statistical analysis of diffusion tensor data

Cited in

(8)

• Statistics for data with geometric structure. Abstracts from the workshop held January 21--27, 2018
• Uniqueness questions in a scaling-rotation geometry on the space of symmetric positive-definite matrices
• Scaling-rotation distance and interpolation of symmetric positive-definite matrices
• Bures–Wasserstein Minimizing Geodesics between Covariance Matrices of Different Ranks
• Stochastic modelling of symmetric positive definite material tensors
• Dynamic principal component analysis from a global perspective
• Eigengap sparsity for covariance parsimony
• Averaging symmetric positive-definite matrices on the space of eigen-decompositions

This page was built for publication: Geometric foundations for scaling-rotation statistics on symmetric positive definite matrices: minimal smooth scaling-rotation curves in low dimensions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q521342)