On the Schoenberg Transformations in Data Analysis: Theory and Illustrations

@article{Bavaud2011OnTS,
  title={On the Schoenberg Transformations in Data Analysis: Theory and Illustrations},
  author={François Bavaud},
  journal={Journal of Classification},
  year={2011},
  volume={28},
  pages={297-314}
}
  • F. Bavaud
  • Published 1 April 2010
  • Mathematics, Computer Science
  • Journal of Classification
The class of Schoenberg transformations, embedding Euclidean distances into higher dimensional Euclidean spaces, is presented, and derived from theorems on positive definite and conditionally negative definite matrices. Original results on the arc lengths, angles and curvature of the transformations are proposed, and visualized on artificial data sets by classical multidimensional scaling. A distance-based discriminant algorithm and a robust multidimensional centroid estimate illustrate the… Expand
Robust Estimation through Schoenberg transformations
Schoenberg transformations, mapping Euclidean configurations into Euclidean configurations, define in turn a transformed inertia, whose minimization produces robust location estimates. The procedureExpand
Euclidean Distances, soft and spectral Clustering on Weighted Graphs
We define a class of Euclidean distances on weighted graphs, enabling to perform thermodynamic soft graph clustering. The class can be constructed form the "raw coordinates" encountered in spectralExpand
Euclidean Distances, Soft and Spectral Clustering on Weighted Graphs
  • F. Bavaud
  • Computer Science, Mathematics
  • ECML/PKDD
  • 2010
TLDR
A class of Euclidean distances on weighted graphs is defined, enabling to perform thermodynamic soft graph clustering, and can be extended by means of higher-dimensional embeddings (Schoenberg transformations). Expand
Relaxed Exponential Kernels for Unsupervised Learning
TLDR
This paper relaxes the global Gaussian assumption made by the Euclidean distance, and proposes a locale Gaussian modelling for the immediate neighbourhood of the samples, resulting in an augmented data space formed by the parameters of the local Gaussians. Expand
Euclidean Distance Matrices (EDMs)
This chapter provides an introduction to Euclidean distance matrices (EDMs). Our primary focus is on various characterizations and basic properties of EDMs. The chapter also discusses methods toExpand
A Mathematical Introduction to Data Science
TLDR
This monograph aims to provide graduate students or senior graduates in applied mathematics, computer science and statistics an introduction to data science from a mathematical perspective, with a divergence to some mathematical theories for deeper understanding. Expand
Computing the Rao ’ s distance between negative binomial distributions . Application to Exploratory Data Analysis
The statistical analysis of counts of living organisms brings information about the collective behavior of species (schooling, habitat preference, etc), possibly depending on their biologicalExpand
On Rayleigh-type formulas for a non-local boundary value problem associated with an integral operator commuting with the Laplacian
Abstract In this article we prove the existence, uniqueness, and simplicity of a negative eigenvalue for a class of integral operators whose kernel is of the form | x − y | ρ , 0 ρ ≤ 1 , x , y ∈ [ −Expand
Graph Drawing by Classical Multidimensional Scaling: New Perspectives
TLDR
This work demonstrates that the classical multidimensional scaling method can be flexibly adapted to provide focus+context layouts and proposes an alternative instantiation that appears to be more suitable for graph drawing and prevents certain degeneracies. Expand
Kernel Mean Embedding of Distributions: A Review and Beyonds
TLDR
A comprehensive review of existing work and recent advances in the Hilbert space embedding of distributions, and to discuss the most challenging issues and open problems that could lead to new research directions. Expand
...
1
2
3
...

References

SHOWING 1-10 OF 63 REFERENCES
On a Connection between Kernel PCA and Metric Multidimensional Scaling
TLDR
The kernel PCA algorithm of Schölkopf, Smola, and Müller can be interpreted as a form of metric multidimensional scaling (MDS) when the kernel function k(x, y) is isotropic, i.e. it depends only on ‖x − y‖. Expand
Spectral Clustering and Multidimensional Scaling: A Unified View
  • F. Bavaud
  • Computer Science, Mathematics
  • Data Science and Classification
  • 2006
TLDR
This paper shows how F and K can be related to each other through a linear or even non-linear transformation leaving the eigen-vectors invariant, and to elucidate the "flow-induced” nature of spatial auto-covariances. Expand
Convolution kernels on discrete structures
We introduce a new method of constructing kernels on sets whose elements are discrete structures like strings, trees and graphs. The method can be applied iteratively to build a kernel on a innniteExpand
Discussion of a set of points in terms of their mutual distances
Necessary and sufficient conditions are given for a set of numbers to be the mutual distances of a set of real points in Euclidean space, and matrices are found whose ranks determine the dimension ofExpand
The Kernel Trick for Distances
A method is described which, like the kernel trick in support vector machines (SVMs), lets us generalize distance-based algorithms to operate in feature spaces, usually nonlinearly related to theExpand
Some distance properties of latent root and vector methods used in multivariate analysis
SUMMARY This paper is concerned with the representation of a multivariate sample of size n as points P1, P2, ..., PI in a Euclidean space. The interpretation of the distance A(Pi, Pj) between the ithExpand
On linear separability of data sets in feature space
TLDR
A method is developed to construct a mapping that maps original data set into a high dimensional feature space, on which inner product is defined by a dot product kernel, and a meaningful formula is obtained to judge linear separability of two infinite data sets in feature space by information in original input space. Expand
Kernel methods in machine learning
We review machine learning methods employing positive definite kernels. These methods formulate learning and estimation problems in a reproducing kernel Hilbert space (RKHS) of functions defined onExpand
Aggregation invariance in general clustering approaches
  • F. Bavaud
  • Computer Science, Mathematics
  • Adv. Data Anal. Classif.
  • 2009
TLDR
Cluster stability and aggregation-invariance of the optimal memberships associated to the various clustering schemes are examined. Expand
Infinitely Divisible Matrices
  • R. Bhatia
  • Mathematics, Computer Science
  • Am. Math. Mon.
  • 2006
TLDR
The general theorems and methods presented in the context of these examples are, in fact, powerful techniques that could be used elsewhere. Expand
...
1
2
3
4
5
...