main concepts scientists sections
SimplePopularAdvanced

How to Thin a Network Without Tearing It express

Original: "Distance Backbones Optimize Spreading Dynamics and Centrality Ranks in the Sparsification of Complex Networks"
· Bernardo Pereira, Felipe Xavier Costa, Luís M. Rocha
arXiv:2603.14564 · 2026-03-15 · CC BY-SA 4.0 · 1 min
Physics and Society cs.SI math.GN
Scientists have found a way to remove redundant links in complex networks while preserving the most important pathways.

A dense social network: everyone is connected to almost everyone else, and in this web it's hard to see who really matters. Too many threads — computers choke, simulations stall. Scientists have figured out how to prune such networks without losing what's essential.

The idea is to remove connections that don't participate in shortest paths. Imagine a city where you keep only the streets that are actually used, demolishing dead ends and redundant alleys. This 'skeleton' of the network preserves all important routes, as well as community structure and the dynamics of information or disease spread.

Here's how it works: each link gets a label — the thinning level at which it disappears. It's like peeling an onion: you remove layer after layer until only the toughest core remains.

They tested it on real social contact networks — from schools, hospitals, and workplaces. It turned out that the variant that best preserves people's importance rankings and spreading dynamics uses a path measure defined by the formula g(x,y) = (\sqrt[3]{x}+\sqrt[3]{y})^3. This skeleton removes more than half of the links, yet the network keeps functioning as it should.

The result is not just a trimmed copy but an accurate map of the most heavily used pathways. This helps speed up epidemic models, find key players in social graphs, and understand where the network is fragile — where it's thin, it tears.

Whole graph →

In some real networks, after thinning, less than half of the links remained, yet the shortest paths between all pairs of nodes were fully preserved.

g(x, y) = (\sqrt[3]{x} + \sqrt[3]{y})^3
A measure of path length between two points, where x and y are edge lengths. Used to define the distance skeleton that best preserves network properties.
Scientists
Claude Shannon
Concepts
weight Graphs graph neural network network dimensionality reduction data compression neural networks in spatial mapping least squares fitting synapse coarse graining bootstrap least squares bayesian hierarchical model redundancy regularization methods phylogenetic network Braess's paradox in biology importance sampling