Recent works have investigated the role of graph bottlenecks in preventing long range information propagation in message-passing graph neural networks, causing the so-called ‘over-squashing’ phenomenon. As a remedy, graph rewiring mech anisms have been proposed as preprocessing steps. Graph Echo State Networks (GESNs) are a reservoir computing model for graphs, where node embeddings are recursively computed by an untrained message-passing function. In this paper, we show that GESNs can achieve a significantly better accuracy on six heterophilic node classification tasks without altering the graph connectivity, thus suggesting a different route for addressing the over-squashing problem.
Leave Graphs Alone: Addressing Over-Squashing without Rewiring
Tortorella, D.;Micheli, A.
2022-01-01
Abstract
Recent works have investigated the role of graph bottlenecks in preventing long range information propagation in message-passing graph neural networks, causing the so-called ‘over-squashing’ phenomenon. As a remedy, graph rewiring mech anisms have been proposed as preprocessing steps. Graph Echo State Networks (GESNs) are a reservoir computing model for graphs, where node embeddings are recursively computed by an untrained message-passing function. In this paper, we show that GESNs can achieve a significantly better accuracy on six heterophilic node classification tasks without altering the graph connectivity, thus suggesting a different route for addressing the over-squashing problem.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
