14 November 2025
Graphs are expressive modeling tools for representing and understanding high-dimensional relationships in data. For example, they allow us to understand the spread of misinformation in social networks or help us learn how drugs interact with the body. Finding communities of similar nodes in graphs, sometimes known as node clustering, is a challenging task with high practical implications as it allows to identify patterns and regularities in highly complex relational data. This is a task particularly challenging in heterophilic graphs, where nodes sharing similarities and membership to the same community are typically distantly connected.
To this end, EMERGE partners from the University of Pisa introduce the Unsupervised Antisymmetric Graph Neural Network (uAGNN), a novel unsupervised community detection approach leveraging non-dissipative dynamical systems to ensure stability and to propagate long-range information effectively. By employing antisymmetric weight matrices, uAGNN captures both local and global graph structures, overcoming the limitations posed by heterophilic scenarios. Extensive experiments across ten datasets demonstrate uAGNN’s superior performance in high and medium heterophilic settings, where traditional methods fail to exploit long-range dependencies. These results highlight uAGNN’s potential as a powerful tool for unsupervised community detection in diverse graph environments.
Read the paper in the link below.
More Information
