Local-to-global Perspectives on Graph Neural Networks

Chen Cai
6/11/2023
cs.LGstat.ML

No official code found. Best implementation: Chen-Cai-OSU/MPNN-GT-Connection (21 stars, last commit 2023-06-13).

Abstract

This thesis presents a local-to-global perspective on graph neural networks (GNN), the leading architecture to process graph-structured data. After categorizing GNN into local Message Passing Neural Networks (MPNN) and global Graph transformers, we present three pieces of work: 1) study the convergence property of a type of global GNN, Invariant Graph Networks, 2) connect the local MPNN and global Graph Transformer, and 3) use local MPNN for graph coarsening, a standard subroutine used in global modeling.

View on arXivView PDF

Code Implementations(1)

Chen-Cai-OSU/MPNN-GT-ConnectionCommunity100%
212Shell, PythonMay 8, 20233 years ago

Similar Papers

Global-Local Graph Neural Networks for Node-Classification

85% match2024

GNN-LoFI: a Novel Graph Neural Network through Localized Feature-based Histogram Intersection

84% match2024

Graphs Unveiled: Graph Neural Networks and Graph Generation

84% match2024

Graph in Graph Neural Network

84% match1 repo162024

A Model-Agnostic Graph Neural Network for Integrating Local and Global Information

84% match2023

Survey on Generalization Theory for Graph Neural Networks

84% match2025

Cite this paper

@article{cai2023localtoglobal,
  title  = {Local-to-global Perspectives on Graph Neural Networks},
  author = {Chen Cai},
  year   = {2023},
  eprint = {2306.06547},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2306.06547v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2306.06547.svg)](https://codeofpaper.com/paper/2306.06547)

Discussion