Principal Sub-manifolds

TL;DR

This paper introduces principal sub-manifolds as a new geometric method for capturing complex, non-linear variation in data lying on Riemannian manifolds, extending PCA to higher dimensions.

Contribution

It extends the concept of principal flows to higher-dimensional sub-manifolds, enabling analysis of more complex data structures on nonlinear manifolds.

Findings

Principal sub-manifolds generalize principal flows to higher dimensions.

The method captures non-geodesic variation in manifold-valued data.

Application demonstrated in shape analysis.

Abstract

We propose a novel method of finding principal components in multivariate data sets that lie on an embedded nonlinear Riemannian manifold within a higher-dimensional space. Our aim is to extend the geometric interpretation of PCA, while being able to capture non-geodesic modes of variation in the data. We introduce the concept of a principal sub-manifold, a manifold passing through a reference point, and at any point on the manifold extending in the direction of highest variation in the space spanned by the eigenvectors of the local tangent space PCA. Compared to recent work for the case where the sub-manifold is of dimension one Panaretos et al. (2014)$-$essentially a curve lying on the manifold attempting to capture one-dimensional variation$-$the current setting is much more general. The principal sub-manifold is therefore an extension of the principal flow, accommodating to capture higher dimensional variation in the data. We show the principal sub-manifold yields the ball spanned by the usual principal components in Euclidean space. By means of examples, we illustrate how to find, use and interpret a principal sub-manifold and we present an application in shape analysis.