Sample-limited L p Barycentric Subspace Analysis on Constant Curvature Spaces - Archive ouverte HAL Access content directly
Conference Papers Year : 2017

Sample-limited L p Barycentric Subspace Analysis on Constant Curvature Spaces

(1)
1

Abstract

Generalizing Principal Component Analysis (PCA) to man-ifolds is pivotal for many statistical applications on geometric data. We rely in this paper on barycentric subspaces, implicitly defined as the lo-cus of points which are weighted means of k + 1 reference points [8, 9]. Barycentric subspaces can naturally be nested and allow the construction of inductive forward or backward nested subspaces approximating data points. We can also consider the whole hierarchy of embedded barycen-tric subspaces defined by an ordered series of points in the manifold (a flag of affine spans): optimizing the accumulated unexplained variance (AUV) over all the subspaces actually generalizes PCA to non Euclidean spaces, a procedure named Barycentric Subspaces Analysis (BSA). In this paper, we first investigate sample-limited inference algorithms where the optimization is limited to the actual data points: this transforms a general optimization into a simple enumeration problem. Second, we propose to robustify the criterion by considering the unexplained p-variance of the residuals instead of the classical 2-variance. This construction is very natural with barycentric subspaces since the affine span is stable under the choice of the value of p. The proposed algorithms are illustrated on examples in constant curvature spaces: optimizing the (accumulated) unexplained p-variance (Lp PBS and BSA) for 0 < p ≤ 1 can identify reference points in clusters of a few points within a large number of random points in spheres and hyperbolic spaces.
Fichier principal
Vignette du fichier
62-Pennec.pdf (1.59 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01574895 , version 1 (16-08-2017)

Identifiers

Cite

Xavier Pennec. Sample-limited L p Barycentric Subspace Analysis on Constant Curvature Spaces. Geometric Sciences of Information (GSI 2017), Nov 2017, Paris, France. pp.20-28, ⟨10.1007/978-3-319-68445-1_3⟩. ⟨hal-01574895⟩
179 View
174 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More