Natural Graph Networks

07/16/2020
by   Pim de Haan, et al.
179

Conventional neural message passing algorithms are invariant under permutation of the messages and hence forget how the information flows through the network. Studying the local symmetries of graphs, we propose a more general algorithm that uses different kernels on different edges, making the network equivariant to local and global graph isomorphisms and hence more expressive. Using elementary category theory, we formalize many distinct equivariant neural networks as natural networks, and show that their kernels are 'just' a natural transformation between two functors. We give one practical instantiation of a natural network on graphs which uses a equivariant message network parameterization, yielding good performance on several benchmarks.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/11/2023

Local-to-global Perspectives on Graph Neural Networks

This thesis presents a local-to-global perspective on graph neural netwo...
research
01/04/2022

DigNet: Digging Clues from Local-Global Interactive Graph for Aspect-level Sentiment Classification

In aspect-level sentiment classification (ASC), state-of-the-art models ...
research
08/07/2018

Message Passing Graph Kernels

Graph kernels have recently emerged as a promising approach for tackling...
research
04/30/2022

Graph Anisotropic Diffusion

Traditional Graph Neural Networks (GNNs) rely on message passing, which ...
research
11/23/2021

Local Permutation Equivariance For Graph Neural Networks

In this work we develop a new method, named locally permutation-equivari...
research
01/07/2018

Covariant Compositional Networks For Learning Graphs

Most existing neural networks for learning graphs address permutation in...
research
02/21/2022

1-WL Expressiveness Is (Almost) All You Need

It has been shown that a message passing neural networks (MPNNs), a popu...

Please sign up or login with your details

Forgot password? Click here to reset