On the Surprising Behaviour of node2vec

TL;DR

This paper investigates the stability and robustness of node2vec graph embeddings, revealing their sensitivity to parameter choices and proposing practical strategies to improve their reliability.

Contribution

It provides a comprehensive analysis of node2vec's embedding stability and introduces methods to enhance robustness against noise and parameter variations.

Findings

Embedding quality varies significantly with parameter choices.

Strategies can improve the stability and robustness of node2vec embeddings.

Analysis highlights the importance of parameter tuning for reliable graph representations.

Abstract

Graph embedding techniques are a staple of modern graph learning research. When using embeddings for downstream tasks such as classification, information about their stability and robustness, i.e., their susceptibility to sources of noise, stochastic effects, or specific parameter choices, becomes increasingly important. As one of the most prominent graph embedding schemes, we focus on node2vec and analyse its embedding quality from multiple perspectives. Our findings indicate that embedding quality is unstable with respect to parameter choices, and we propose strategies to remedy this in practice.