Building powerful and equivariant graph neural networks with structural message-passing

C Vignac, A Loukas, P Frossard - Advances in neural …, 2020 - proceedings.neurips.cc
Message-passing has proved to be an effective way to design graph neural networks, as it is
able to leverage both permutation equivariance and an inductive bias towards learning local …

Theory of graph neural networks: Representation and learning

S Jegelka - The International Congress of Mathematicians, 2022 - ems.press
Abstract Graph Neural Networks (GNNs), neural network architectures targeted to learning
representations of graphs, have become a popular learning model for prediction tasks on …

Convergence and stability of graph convolutional networks on large random graphs

N Keriven, A Bietti, S Vaiter - Advances in Neural …, 2020 - proceedings.neurips.cc
We study properties of Graph Convolutional Networks (GCNs) by analyzing their behavior
on standard models of random graphs, where nodes are represented by random latent …

How hard is to distinguish graphs with graph neural networks?

A Loukas - Advances in neural information processing …, 2020 - proceedings.neurips.cc
A hallmark of graph neural networks is their ability to distinguish the isomorphism class of
their inputs. This study derives hardness results for the classification variant of graph …

On the universality of graph neural networks on large random graphs

N Keriven, A Bietti, S Vaiter - Advances in Neural …, 2021 - proceedings.neurips.cc
We study the approximation power of Graph Neural Networks (GNNs) on latent position
random graphs. In the large graph limit, GNNs are known to converge to …

A graph similarity for deep learning

S Ok - Advances in Neural Information Processing Systems, 2020 - proceedings.neurips.cc
Graph neural networks (GNNs) have been successful in learning representations from
graphs. Many popular GNNs follow the pattern of aggregate-transform: they aggregate the …

Convergence of Message Passing Graph Neural Networks with Generic Aggregation On Large Random Graphs

M Cordonnier, N Keriven, N Tremblay… - arXiv preprint arXiv …, 2023 - arxiv.org
We study the convergence of message passing graph neural networks on random graph
models to their continuous counterpart as the number of nodes tends to infinity. Until now …

Statistical Guarantees for Link Prediction using Graph Neural Networks

A Chung, A Saberi, M Austern - arXiv preprint arXiv:2402.02692, 2024 - arxiv.org
This paper derives statistical guarantees for the performance of Graph Neural Networks
(GNNs) in link prediction tasks on graphs generated by a graphon. We propose a linear …

Fundamental limits of deep graph convolutional networks for graph classification

A Magner, M Baranwal, AO Hero - IEEE transactions on …, 2022 - ieeexplore.ieee.org
Graph convolutional networks (GCNs) are a widely used method for graph representation
learning. To elucidate their capabilities and limitations for graph classification, we …

Fundamental limits of deep graph convolutional networks

A Magner, M Baranwal, AO Hero III - arXiv preprint arXiv:1910.12954, 2019 - arxiv.org
Graph convolutional networks (GCNs) are a widely used method for graph representation
learning. To elucidate the capabilities and limitations of GCNs, we investigate their power …