Least squares approximations of measures via geometric condition numbers

Abstract

For a probability measure on a real separable Hilbert space, we are interested in "volume-based" approximations of the d-dimensional least squares error of it, i.e., least squares error with respect to a best fit d-dimensional affine subspace. Such approximations are given by averaging real-valued multivariate functions which are typically scalings of squared (d+1)-volumes of (d+1)-simplices. Specifically, we show that such averages are comparable to the square of the d-dimensional least squares error of that measure, where the comparison depends on a simple quantitative geometric property of it. This result is a higher dimensional generalization of the elementary fact that the double integral of the squared distances between points is proportional to the variance of measure. We relate our work to two recent algorithms, one for clustering affine subspaces and the other for Monte-Carlo SVD based on volume sampling.

References in corpus (5)

• Spectral clustering based on local linear approximations
• Foundations of a Multi-way Spectral Clustering Framework for Hybrid Linear Modeling
• Integral Menger curvature for surfaces
• On d-dimensional d-Semimetrics and Simplex-Type Inequalities for High-Dimensional Sine Functions
• High-Dimensional Menger-Type Curvatures-Part II: d-Separation and a Menagerie of Curvatures

Cited by in corpus (2)

• Menger curvatures and $C^{1,\alpha }$ rectifiability of measures
• Estimating discrete curvatures in terms of beta numbers