Statistical Analysis and Parameter Selection for Mapper
Computational Geometry
2017-11-10 v2 Algebraic Topology
Methodology
Abstract
In this article, we study the question of the statistical convergence of the 1-dimensional Mapper to its continuous analogue, the Reeb graph. We show that the Mapper is an optimal estimator of the Reeb graph, which gives, as a byproduct, a method to automatically tune its parameters and compute confidence regions on its topological features, such as its loops and flares. This allows to circumvent the issue of testing a large grid of parameters and keeping the most stable ones in the brute-force setting, which is widely used in visualization, clustering and feature selection with the Mapper.
Cite
@article{arxiv.1706.00204,
title = {Statistical Analysis and Parameter Selection for Mapper},
author = {Mathieu Carrière and Bertrand Michel and Steve Oudot},
journal= {arXiv preprint arXiv:1706.00204},
year = {2017}
}
Comments
Minor modifications