An Infinitesimal Probabilistic Model for Principal Component Analysis of Manifold Valued Data

01/31/2018
by   Stefan Sommer, et al.
0

We provide a probabilistic and infinitesimal view of how the principal component analysis procedure (PCA) can be generalized to analysis of nonlinear manifold valued data. Starting with the probabilistic PCA interpretation of the Euclidean PCA procedure, we show how PCA can be generalized to manifolds in an intrinsic way that does not resort to linearization of the data space. The underlying probability model is constructed by mapping a Euclidean stochastic process to the manifold using stochastic development of Euclidean semimartingales. The construction uses a connection and bundles of covariant tensors to allow global transport of principal eigenvectors, and the model is thereby an example of how principal fiber bundles can be used to handle the lack of global coordinate system and orientations that characterizes manifold valued statistics. We show how curvature implies non-integrability of the equivalent of Euclidean principal subspaces, and how the stochastic flows provide an alternative to explicit construction of such subspaces. We describe two estimation procedures for inference of parameters and prediction of principal components, and we give examples of properties of the model on surfaces embedded in ^3.

READ FULL TEXT

page 15

page 16

research
08/13/2019

Principal symmetric space analysis

We develop a novel analogue of Euclidean PCA (principal component analys...
research
09/05/2022

A geometric framework for asymptotic inference of principal subspaces in PCA

In this article, we develop an asymptotic method for testing hypothesis ...
research
01/06/2023

Principal Component Analysis in Space Forms

Principal component analysis (PCA) is a workhorse of modern data science...
research
12/12/2018

Modeling Longitudinal Data on Riemannian Manifolds

When considering functional principal component analysis for sparsely ob...
research
06/18/2021

Curvature of point clouds through principal component analysis

In this article, we study curvature-like feature value of data sets in E...
research
08/16/2016

Parameterized Principal Component Analysis

When modeling multivariate data, one might have an extra parameter of co...
research
02/06/2015

A Fingerprint-based Access Control using Principal Component Analysis and Edge Detection

This paper presents a novel approach for deciding on the appropriateness...

Please sign up or login with your details

Forgot password? Click here to reset