Learning low-rank latent mesoscale structures in networks. (arXiv:2102.06984v3 [cs.SI] UPDATED)

It is common to use networks to encode the architecture of interactions
between entities in complex systems in applications in the physical,
biological, social, and information sciences. To study the large-scale behavior
of complex systems, it is useful to study mesoscale structures in networks as
building blocks that influence such behavior. We present a new approach for
describing low-rank mesoscale structure in networks, and we illustrate our
approach using several synthetic network models and empirical friendship,
collaboration, and protein--protein interaction …

arxiv learning networks