P-Tensors: a General Formalism for Constructing Higher Order Message Passing Networks

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Hands, Andrew, Sun, Tianyi, Kondor, Risi
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917245933846528
author Hands, Andrew
Sun, Tianyi
Kondor, Risi
author_facet Hands, Andrew
Sun, Tianyi
Kondor, Risi
contents Several recent papers have proposed increasing the expressive power of graph neural networks by exploiting subgraphs or other topological structures. In parallel, researchers have investigated higher order permutation equivariant networks. In this paper we tie these two threads together by providing a general framework for higher order permutation equivariant message passing in subgraph neural networks. In this paper we introduce a new type of mathematical object called $P$-tensors, which provide a simple way to define the most general form of permutation equivariant message passing in both the above two categories of networks. We show that the P-Tensors paradigm can achieve state-of-the-art performance on benchmark molecular datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2306_10767
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle P-Tensors: a General Formalism for Constructing Higher Order Message Passing Networks
Hands, Andrew
Sun, Tianyi
Kondor, Risi
Machine Learning
Statistics Theory
Several recent papers have proposed increasing the expressive power of graph neural networks by exploiting subgraphs or other topological structures. In parallel, researchers have investigated higher order permutation equivariant networks. In this paper we tie these two threads together by providing a general framework for higher order permutation equivariant message passing in subgraph neural networks. In this paper we introduce a new type of mathematical object called $P$-tensors, which provide a simple way to define the most general form of permutation equivariant message passing in both the above two categories of networks. We show that the P-Tensors paradigm can achieve state-of-the-art performance on benchmark molecular datasets.
title P-Tensors: a General Formalism for Constructing Higher Order Message Passing Networks
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2306.10767