New paper on arXiv: Exponential random graphs behave like mixtures of stochastic block models

In recent years, there has been an increasing interest in academic papers whose opening paragraphs describe the increasing interest in the (increasingly common) field of network theory. From mathematicians to sociologists, many a great scientist have written introductory paragraphs on this interesting explosion. For example:

Chatterjee and Diaconis (2013):

Bhamidi, Bresler and Sly (2011):

Yan, Leng and Zhu (2016):

Shalizi and Rinaldo (2013):

and many others.

I am happy to announce with utmost pride that my advisor Ronen Eldan and I have joined the ranks, and have also written an introductory paragraph describing the increasing interest in large networks:

You can read the entire paper here: Exponential random graphs behave like mixtures of stochastic block models. In short, we show that exponential random graphs, which are a common way of generating graphs with dependence between the edges, behave like a mixture of graphs with independent edges. The way those independent edges behave is determined by a matrix equation; we show that in the general case the solutions to this equation are always close block matrices, and we also solve the equation explicitly for some particular cases. This result is so far the strongest known characterization of the exponential random graphs.

A more complete post will come in the near future (in geological timescales, anyway).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s