Graph neural network outputs are almost surely asymptotically constant

arXiv:2403.03880v1 Announce Type: new
Abstract: Graph neural networks (GNNs) are the predominant architectures for a variety of learning tasks on graphs. We present a new angle on the expressive power of GNNs by studying how the predictions of a GNN probabilistic classifier evolve as we apply it on larger graphs drawn from some random graph model. We show that the output converges to a constant function, which upper-bounds what these classifiers can express uniformly. This convergence phenomenon applies to a …

abstract apply architectures arxiv classifier cs.lg cs.lo gnn gnns graph graph neural network graph neural networks graphs network networks neural network neural networks power predictions studying tasks type