Geometry and clustering with metrics derived from separable Bregman divergences
Machine Learning
2018-10-26 v1 Machine Learning
Abstract
Separable Bregman divergences induce Riemannian metric spaces that are isometric to the Euclidean space after monotone embeddings. We investigate fixed rate quantization and its codebook Voronoi diagrams, and report on experimental performances of partition-based, hierarchical, and soft clustering algorithms with respect to these Riemann-Bregman distances.
Cite
@article{arxiv.1810.10770,
title = {Geometry and clustering with metrics derived from separable Bregman divergences},
author = {Erika Gomes-Gonçalves and Henryk Gzyl and Frank Nielsen},
journal= {arXiv preprint arXiv:1810.10770},
year = {2018}
}
Comments
23 pages