Last week NetSI hosted the third and final Future of Network Science Summit & Collabathon (March 3-7), a week-long event where junior scholars from US and European research institutions and universities worked to document, compare, and advance methods for coarse-graining complex networks.
The Summit & Collabathon are part of the AccelNet-MultiNet program coordinated jointly with Indiana University and the Network Science Institute at Northeastern University and supported by the National Science Foundation, with the goal to form strong and lasting collaborations around core topics in multi-layer and higher-order networks. This scientific focus, which is at the heart of the AccelNet-MultiNet program, is common to virtually all methodological and theoretical considerations across different network applications. The last thirty years of network science have witnessed formidable advances in our understanding of the emergence of networked structures and of the processes occurring on networks. While it has always been understood that network systems do not live in isolation, it is only in recent years that we have the mathematical capabilities and data availability to rigorously explore the nature of ‘networks of networks’. Multi-layer network science refers to the set of tools and theories used to model interactions between multiple networks, or layers of networks.
To continue advancing the research in this area of network science and creatively tackle its scientific challenges, Brennan Klein, Assistant Teaching Professor and faculty member of the Network Science Institute at Northeastern University designed the Collabathon (combination of “collaboration” and “hackathon”) as a way to bring together knowledge and skills across a group of researchers during a multi-day event.
Building on the progress made during last year’s Collabathon in Zaragoza, Spain—where attendees significantly advanced CoarseNet, a Python package for comparing and evaluating techniques for deriving higher-scale, coarse-grained representations of networks—this year’s Collabathon focused on organizing these techniques into a taxonomy that enables better comparison and potential consolidation of methods. The key question driving this project was: What is the right scale to model a complex system, from cells to cities, what we can learn from different modeling approaches and how to determine which is the most effective approach.
The results of the Collabathon and the Summit conclusions will be published in a forthcoming final report, contributing to new perspectives that will advance research and discovery in the field of multilayer network science, and establish new capabilities for international collaborations.
