11 April 2025
A common problem in Message-Passing Neural Networks is oversquashing, the limited ability to facilitate effective information flow between distant nodes. Oversquashing is attributed to the exponential decay in information transmission as node distances increase. This paper introduces a novel perspective to address oversquashing, leveraging dynamical systems properties of global and local non-dissipativity, that enable the maintenance of a constant information flow rate.
In this work, EMERGE partners from the University of Pisa present SWAN, a uniquely parameterized GNN model with antisymmetry both in space and weight domains, in order to obtain non-dissipativity. Their theoretical analysis asserts that by implementing these properties, SWAN offers an enhanced ability to transmit information over extended distances. Empirical evaluations on synthetic and real-world benchmarks that emphasize long-range interactions validate the theoretical understanding of SWAN, and its ability to mitigate oversquashing.
Read the paper in the link below.
More Information
