Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Adam-Day, Sam, Benedikt, Michael, Ceylan, İsmail İlkan, Finkelshtein, Ben
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2403.03880
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917830853656576
author Adam-Day, Sam
Benedikt, Michael
Ceylan, İsmail İlkan
Finkelshtein, Ben
author_facet Adam-Day, Sam
Benedikt, Michael
Ceylan, İsmail İlkan
Finkelshtein, Ben
contents We present a new angle on the expressive power of graph neural networks (GNNs) by studying how the predictions of real-valued GNN classifiers, such as those classifying graphs probabilistically, evolve as we apply them 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 uniformly express. This strong convergence phenomenon applies to a very wide class of GNNs, including state of the art models, with aggregates including mean and the attention-based mechanism of graph transformers. Our results apply to a broad class of random graph models, including sparse and dense variants of the Erdős-Rényi model, the stochastic block model, and the Barabási-Albert model. We empirically validate these findings, observing that the convergence phenomenon appears not only on random graphs but also on some real-world graphs.
format Preprint
id arxiv_https___arxiv_org_abs_2403_03880
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Almost Surely Asymptotically Constant Graph Neural Networks
Adam-Day, Sam
Benedikt, Michael
Ceylan, İsmail İlkan
Finkelshtein, Ben
Machine Learning
Logic in Computer Science
We present a new angle on the expressive power of graph neural networks (GNNs) by studying how the predictions of real-valued GNN classifiers, such as those classifying graphs probabilistically, evolve as we apply them 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 uniformly express. This strong convergence phenomenon applies to a very wide class of GNNs, including state of the art models, with aggregates including mean and the attention-based mechanism of graph transformers. Our results apply to a broad class of random graph models, including sparse and dense variants of the Erdős-Rényi model, the stochastic block model, and the Barabási-Albert model. We empirically validate these findings, observing that the convergence phenomenon appears not only on random graphs but also on some real-world graphs.
title Almost Surely Asymptotically Constant Graph Neural Networks
topic Machine Learning
Logic in Computer Science
url https://arxiv.org/abs/2403.03880