English

Phase transition in the recoverability of network history

Physics and Society 2019-12-25 v3 Statistical Mechanics Machine Learning

Abstract

Network growth processes can be understood as generative models of the structure and history of complex networks. This point of view naturally leads to the problem of network archaeology: reconstructing all the past states of a network from its structure---a difficult permutation inference problem. In this paper, we introduce a Bayesian formulation of network archaeology, with a generalization of preferential attachment as our generative mechanism. We develop a sequential Monte Carlo algorithm to evaluate the posterior averages of this model, as well as an efficient heuristic that uncovers a history well correlated with the true one, in polynomial time. We use these methods to identify and characterize a phase transition in the quality of the reconstructed history, when they are applied to artificial networks generated by the model itself. Despite the existence of a no-recovery phase, we find that nontrivial inference is possible in a large portion of the parameter space as well as on empirical data.

Keywords

Cite

@article{arxiv.1803.09191,
  title  = {Phase transition in the recoverability of network history},
  author = {Jean-Gabriel Young and Guillaume St-Onge and Edward Laurence and Charles Murphy and Laurent Hébert-Dufresne and Patrick Desrosiers},
  journal= {arXiv preprint arXiv:1803.09191},
  year   = {2019}
}

Comments

18 pages, 10 figures. Supplemental Material available online at https://journals.aps.org/prx/supplemental/10.1103/PhysRevX.9.041056/jgy2019_prx_SM.pdf