Unsupervised Learning of 3-colorings using Simplicial Higher-Order Neural Networks

Lucas Laird, Robin Walters, Wolfgang Gatterbauer

Keywords: GNNs, Higher-Order Networks (HONs), Constraint Satisfaction Problems (CSPs), Simplicial Complexe

TL;DR: We show that Higher-Order Networks are a promising route towards solving historically challenging problems for GNNs such as 3-coloring.

Abstract: We propose Higher-Order Networks (HONs) for historically challenging problems for Graph Neural Networks (GNNs), such as Constraint Satisfaction Problems (CSPs). We apply a simple extension of GNNs to HONs and show its advantages for solving 3-coloring.

Type Of Submission: Extended Abstract (4 pages, non-archival)

Submission Number: 80