English

Power Weighted Shortest Paths for Clustering Euclidean Data

Machine Learning 2019-09-05 v3 Machine Learning

Abstract

We study the use of power weighted shortest path distance functions for clustering high dimensional Euclidean data, under the assumption that the data is drawn from a collection of disjoint low dimensional manifolds. We argue, theoretically and experimentally, that this leads to higher clustering accuracy. We also present a fast algorithm for computing these distances.

Keywords

Cite

@article{arxiv.1905.13345,
  title  = {Power Weighted Shortest Paths for Clustering Euclidean Data},
  author = {Daniel Mckenzie and Steven Damelin},
  journal= {arXiv preprint arXiv:1905.13345},
  year   = {2019}
}

Comments

24 pages. Final version. To appear in Foundations of Data Science

R2 v1 2026-06-23T09:34:15.489Z