AI Paper Atlas

Research map / Graph learning / GNN architectures

k-th Order Deep Homomorphism Networks

Takanori Maehara

NeurIPS 2026 · Graph learning › GNN architectures · approach: Substructure-aware GNN expressivity

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Extends Deep Homomorphism Networks to k-labelled patterns, giving a pattern-algebraic framework for higher-order GNN expressivity along skeleton complexity k and pattern vocabulary axes.

Problem

Characterizing the expressivity of higher-order graph neural networks, which depends on both global structures and local structures captured by pattern vocabularies.

Approach

Extends rooted-pattern Deep Homomorphism Networks to k-labelled patterns, separating expressivity into two independent axes (skeleton complexity k for global structure, pattern vocabulary for local structure), and introduces an auxiliary gluing operation to combine patterns.

Experimental setup

Builds on: Deep Homomorphism Networks

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy