Alert button

Spectral clustering under degree heterogeneity: a case for the random walk Laplacian

May 04, 2021
Alexander Modell, Patrick Rubin-Delanchy

Figure 1 for Spectral clustering under degree heterogeneity: a case for the random walk Laplacian
Figure 2 for Spectral clustering under degree heterogeneity: a case for the random walk Laplacian
Figure 3 for Spectral clustering under degree heterogeneity: a case for the random walk Laplacian
Figure 4 for Spectral clustering under degree heterogeneity: a case for the random walk Laplacian

Share this with someone who'll enjoy it:

This paper shows that graph spectral embedding using the random walk Laplacian produces vector representations which are completely corrected for node degree. Under a generalised random dot product graph, the embedding provides uniformly consistent estimates of degree-corrected latent positions, with asymptotically Gaussian error. In the special case of a degree-corrected stochastic block model, the embedding concentrates about K distinct points, representing communities. These can be recovered perfectly, asymptotically, through a subsequent clustering step, without spherical projection, as commonly required by algorithms based on the adjacency or normalised, symmetric Laplacian matrices. While the estimand does not depend on degree, the asymptotic variance of its estimate does -- higher degree nodes are embedded more accurately than lower degree nodes. Our central limit theorem therefore suggests fitting a weighted Gaussian mixture model as the subsequent clustering step, for which we provide an expectation-maximisation algorithm.

* 22 pages, 10 figures  
View paper onarxiv icon

Share this with someone who'll enjoy it: