Fast and scalable likelihood maximization for Exponential Random Graph Models with local constraints

Abstract

Exponential Random Graph Models (ERGMs) have gained increasing popularity over the years. Rooted into statistical physics, the ERGMs framework has been successfully employed for reconstructing networks, detecting statistically significant patterns in graphs, counting networked configurations with given properties. From a technical point of view, the ERGMs workflow is defined by two subsequent optimization steps: the first one concerns the maximization of Shannon entropy and leads to identify the functional form of the ensemble probability distribution that is maximally non-committal with respect to the missing information; the second one concerns the maximization of the likelihood function induced by this probability distribution and leads to its numerical determination. This second step translates into the resolution of a system of O(N) non-linear, coupled equations (with N being the total number of nodes of the network under analysis), a problem that is affected by three main issues, i.e. accuracy, speed and scalability. The present paper aims at addressing these problems by comparing the performance of three algorithms (i.e. Newton’s method, a quasi-Newton method and a recently-proposed fixed-point recipe) in solving several ERGMs, defined by binary and weighted constraints in both a directed and an undirected fashion. While Newton’s method performs best for relatively little networks, the fixed-point recipe is to be preferred when large configurations are considered, as it ensures convergence to the solution within seconds for networks with hundreds of thousands of nodes (e.g. the Internet, Bitcoin). We attach to the paper a Python code implementing the three aforementioned algorithms on all the ERGMs considered in the present work.

Brexit and bots: characterizing the behaviour of automated accounts on Twitter during the UK election

“…We use the implementation proposed in [ 40 ] and the relative code of the package NEMtropy 7 for the computation of the randomization and of the monopartite projection. In the following, the nodes corresponding to one layer will be called with the letter as they correspond to the rows layer in a biadjacency matrix representation of the network, and the opposite layer will be indexed as as they correspond to columns.…”

“…In order to find the numerical value of the probability, we set the average degrees of the model to the observed ones and solve the set of nonlinear equations where and are the degree of the node r and c respectively, and ∗ indicates the observed empirical value. More details can be found in [ 40 ].…”

Structural complementarity and similarity: linking relational principles to network motifs

“…Observed values of structural coefficients were calibrated based on Undirected Binary Configuration Model (UBCM) [52]. The model induces a maximum entropy probability distribution over undirected and unweighted networks with n nodes constrained to have a specific expected degree sequence.…”

Firms' Challenges and Social Responsibilities during Covid-19: a Twitter Analysis

“…For the implementation of the BiCM, we used the recently released python module NEMtropy, presented in [61].…”

“…After the introduction of the BiCM in subsection B.1, we will show in subsection B.2 how the null-model can be used as a benchmark to validate the projection on one of the two layers, as proposed in [23]. Let us finally remark that the exact implementation of the null-model was performed via the python module NEMtropy, presented in [61].…”