Edge-Wise Graph-Instructed Neural Networks

Francesco Della Santa, Antonio Mastropietro, Sandra Pieraccini, Francesco Vaccarino
9/12/2024
cs.LGcs.AImath.NA

No official code found. Best implementation: Fra0013To/GINN (6 stars, MIT, last commit 2026-04-28).

Abstract

The problem of multi-task regression over graph nodes has been recently approached through Graph-Instructed Neural Network (GINN), which is a promising architecture belonging to the subset of message-passing graph neural networks. In this work, we discuss the limitations of the Graph-Instructed (GI) layer, and we formalize a novel edge-wise GI (EWGI) layer. We discuss the advantages of the EWGI layer and we provide numerical evidence that EWGINNs perform better than GINNs over some graph-structured input data, like the ones inferred from the Barabasi-Albert graph, and improve the training regularization on graphs with chaotic connectivity, like the ones inferred from the Erdos-Renyi graph.

View on arXivView PDF

Code Implementations(1)

Fra0013To/GINNCommunity100%
MIT

Graph-Informed Neural Networks

62PythonFeb 28, 20225 months agoMIT

Similar Papers

Sparse Implementation of Versatile Graph-Informed Layers

83% match1 repo62024

Edge Prompt Tuning for Graph Neural Networks

83% match1 repo102025

Revisiting Graph Neural Networks on Graph-level Tasks: Comprehensive Experiments, Analysis, and Improvements

83% match2025

Graph Attention Network for Node Regression on Random Geometric Graphs with Erdős--Rényi contamination

82% match2026

Beyond Message Passing: Neural Graph Pattern Machine

82% match1 repo152025

RegExplainer: Generating Explanations for Graph Neural Networks in Regression Tasks

82% match1 repo92023

Cite this paper

@article{santa2024edgewise,
  title  = {Edge-Wise Graph-Instructed Neural Networks},
  author = {Francesco Della Santa and Antonio Mastropietro and Sandra Pieraccini and Francesco Vaccarino},
  year   = {2024},
  eprint = {2409.08023},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.08023v2}
}

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/2409.08023.svg)](https://codeofpaper.com/paper/2409.08023)

Discussion