Clustering in Hilbert simplex geometry
arXiv:1704.00454 · doi:10.1007/978-3-030-02520-5_11
Abstract
Clustering categorical distributions in the finite-dimensional probability simplex is a fundamental task met in many applications dealing with normalized histograms. Traditionally, the differential-geometric structures of the probability simplex have been used either by (i) setting the Riemannian metric tensor to the Fisher information matrix of the categorical distributions, or (ii) defining the dualistic information-geometric structure induced by a smooth dissimilarity measure, the Kullback-Leibler divergence. In this work, we introduce for clustering tasks a novel computationally-friendly framework for modeling geometrically the probability simplex: The {\em Hilbert simplex geometry}. In the Hilbert simplex geometry, the distance is the non-separable Hilbert's metric distance which satisfies the property of information monotonicity with distance level set functions described by polytope boundaries. We show that both the Aitchison and Hilbert simplex distances are norm distances on normalized logarithmic representations with respect to the and variation norms, respectively. We discuss the pros and cons of those different statistical modelings, and benchmark experimentally these different kind of geometries for center-based -means and -center clustering. Furthermore, since a canonical Hilbert distance can be defined on any bounded convex subset of the Euclidean space, we also consider Hilbert's geometry of the elliptope of correlation matrices and study its clustering performances compared to Fröbenius and log-det divergences.
arXiv comment48 pages
References in corpus (12)
- Inference and Evaluation of the Multinomial Mixture Model for Text Clustering
- On Approximating the Riemannian 1-Center
- Minimum Enclosing Polytope in High Dimensions
- Cramer-Rao Lower Bound and Information Geometry
- On Hölder projective divergences
- Clustering in Hilbert simplex geometry
- No dimension independent Core-Sets for Containment under Homothetics
- Simplicial faces of the set of correlation matrices
- Medians and means in Finsler geometry
- The geometric stability of Voronoi diagrams in normed spaces which are not uniformly convex
- Further heuristics for -means: The merge-and-split heuristic and the -means
- The Epic Story of Maximum Likelihood
Cited by in corpus (6)
- Clustering in Hilbert simplex geometry
- Hilbert geometry of the Siegel disk: The Siegel-Klein disk model
- Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach
- Tropical convexity in location problems
- cCorrGAN: Conditional Correlation GAN for Learning Empirical Conditional Distributions in the Elliptope
- Geodesic of the Quotient-Affine Metrics on Full-Rank Correlation Matrices