Graph Laplacians on Shared Nearest Neighbor graphs and graph Laplacians on $k$-Nearest Neighbor graphs having the same limit

TL;DR

This paper investigates the theoretical properties of Shared Nearest Neighbor (SNN) graphs and their Laplacians, revealing that their asymptotic behavior aligns with that of k-NN graphs, and establishing convergence rates.

Contribution

It provides the first theoretical analysis of SNN graph Laplacians, showing their asymptotic equivalence to k-NN graph Laplacians and deriving convergence rates.

Findings

SNN graph Laplacians have the same continuum limit as k-NN graph Laplacians.

The pointwise convergence rate of the graph Laplacian is linear in (k/n)^{1/m}.

Theoretical foundation for using SNN graphs in high-dimensional data analysis.

Abstract

A Shared Nearest Neighbor (SNN) graph is a type of graph construction using shared nearest neighbor information, which is a secondary similarity measure based on the rankings induced by a primary $k$-nearest neighbor ($k$-NN) measure. SNN measures have been touted as being less prone to the curse of dimensionality than conventional distance measures, and thus methods using SNN graphs have been widely used in applications, particularly in clustering high-dimensional data sets and in finding outliers in subspaces of high dimensional data. Despite this, the theoretical study of SNN graphs and graph Laplacians remains unexplored. In this pioneering work, we make the first contribution in this direction. We show that large scale asymptotics of an SNN graph Laplacian reach a consistent continuum limit; this limit is the same as that of a $k$-NN graph Laplacian. Moreover, we show that the pointwise convergence rate of the graph Laplacian is linear with respect to $(k/n)^{1/m}$ with high probability.