Spectral embedding for dynamic networks with stability guarantees

Ian Gallagher, Andrew Jones, Patrick Rubin-Delanchy

Keywords: graph, dynamic network, stochastic block model, spectral embedding, clustering, time series analysis

Abstract: We consider the problem of embedding a dynamic network, to obtain time-evolving vector representations of each node, which can then be used to describe changes in behaviour of individual nodes, communities, or the entire graph. Given this open-ended remit, we argue that two types of stability in the spatio-temporal positioning of nodes are desirable: to assign the same position, up to noise, to nodes behaving similarly at a given time (cross-sectional stability) and a constant position, up to noise, to a single node behaving similarly across different times (longitudinal stability). Similarity in behaviour is defined formally using notions of exchangeability under a dynamic latent position network model. By showing how this model can be recast as a multilayer random dot product graph, we demonstrate that unfolded adjacency spectral embedding satisfies both stability conditions. We also show how two alternative methods, omnibus and independent spectral embedding, alternately lack one or the other form of stability.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

TL;DR: We show how to embed a dynamic network to allow coherent spatio-temporal clustering

Code: https://github.com/iggallagher/Dynamic-Network-Embedding

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2106.01282/code)