Spectral embedding of weighted graphs
Machine Learning
2023-01-23 v4 Machine Learning
Abstract
When analyzing weighted networks using spectral embedding, a judicious transformation of the edge weights may produce better results. To formalize this idea, we consider the asymptotic behavior of spectral embedding for different edge-weight representations, under a generic low rank model. We measure the quality of different embeddings -- which can be on entirely different scales -- by how easy it is to distinguish communities, in an information-theoretic sense. For common types of weighted graphs, such as count networks or p-value networks, we find that transformations such as tempering or thresholding can be highly beneficial, both in theory and in practice.
Cite
@article{arxiv.1910.05534,
title = {Spectral embedding of weighted graphs},
author = {Ian Gallagher and Andrew Jones and Anna Bertiger and Carey Priebe and Patrick Rubin-Delanchy},
journal= {arXiv preprint arXiv:1910.05534},
year = {2023}
}
Comments
27 pages, 5 figures