Scale and curvature effects in principal geodesic analysis

DOI10.1016/J.JMVA.2016.09.009zbMATH Open1351.60012arXiv1610.01537OpenAlexW2522210570MaRDI QIDQ2374401FDOQ2374401

Authors: Drew Lazar, Lizhen Lin 

Publication date: 15 December 2016

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: There is growing interest in using the close connection between differential geometry and statistics to model smooth manifold-valued data. In particular, much work has been done recently to generalize principal component analysis (PCA), the method of dimension reduction in linear spaces, to Riemannian manifolds. One such generalization is known as principal geodesic analysis (PGA). This paper, in a novel fashion, obtains Taylor expansions in scaling parameters introduced in the domain of objective functions in PGA. It is shown this technique not only leads to better closed-form approximations of PGA but also reveals the effects that scale, curvature and the distribution of data have on solutions to PGA and on their differences to first-order tangent space approximations. This approach should be able to be applied not only to PGA but also to other generalizations of PCA and more generally to other intrinsic statistics on Riemannian manifolds.

Full work available at URL: https://arxiv.org/abs/1610.01537

Recommendations

• An infinitesimal probabilistic model for principal component analysis of manifold valued data
• Principal symmetric space analysis
• Principal component analysis for Riemannian manifolds, with an application to triangular shape spaces
• Principal flows
• Statistical methods generalizing principal component analysis to non-Euclidean spaces

zbMATH Keywords

dimension reductioncurvature effectssymmetric spacesdata scalingdiffusion tensorsmanifold-valued statisticsprincipal geodesic analysis (PGA)

Mathematics Subject Classification ID

Directional data; spatial statistics (62H11) Geometric probability and stochastic geometry (60D05) Geodesics in global differential geometry (53C22)

Cites Work

• Title not available (Why is that?)
• Analysis of principal nested spheres
• Title not available (Why is that?)
• Title not available (Why is that?)
• Optimization
• Geodesic regression and the theory of least squares on Riemannian manifolds
• Title not available (Why is that?)
• Title not available (Why is that?)
• Riemannian center of mass and mollifier smoothing
• Title not available (Why is that?)
• Using Orientation Statistics to Investigate Variations in Human Kinematics
• Derivatives of the matrix exponential and their computation
• Comparison theorems in Riemannian geometry
• Principal component analysis for Riemannian manifolds, with an application to triangular shape spaces
• Title not available (Why is that?)

Cited In (5)

• Long-time principal geodesic analysis in director-based dynamics of hybrid mechanical systems
• A geometric interpretation of Darroch and Ratcliff's generalized iterative scaling
• Robust optimization and inference on manifolds
• Statistical inference on the Hilbert sphere with application to random densities
• A 50-year personal journey through time with principal component analysis

Uses Software

• Rotations

This page was built for publication: Scale and curvature effects in principal geodesic analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2374401)