AI Paper Atlas

Research map / Graph learning / GNN architectures

Sheaf Hypergraph Networks

Iulia Duta, Giulia Cassarà, Fabrizio Silvestri, Pietro Lio

NeurIPS 2023 · Graph learning › GNN architectures · approach: Oversmoothing and heterophily · 9 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace trivial-sheaf Laplacian with learnable linear sheaf Laplacian (HyperGNN -> SheafHyperGNN)HyperGNN · Senateaccuracy (%): 48.59 -> 68.73Improved
replace trivial-sheaf with sheaf in non-linear Laplacian (HyperGCN -> SheafHyperGCN)HyperGCN · Senateaccuracy (%): 51.13 -> 66.33Improved
replace HyperGNN with SheafHyperGNNHyperGNN · Houseaccuracy (%): 61.39 -> 73.84Improved
replace HyperGNN with SheafHyperGNNHyperGNN · Coraaccuracy (%): 79.39 -> 81.30Improved
replace HyperGNN with SheafHyperGNNHyperGNN · synthetic heterophilic alpha=7accuracy (%): 63.8 -> 77.3Improved
use linear sheaf Laplacian (SheafHyperGNN) instead of non-linear (SheafHyperGCN)SheafHyperGNN · Cora_CAaccuracy (%): 83.26 -> 85.52Improved
use general d x d restriction maps instead of diagonal (ablation)SheafHyperGNN · Cora_CAaccuracy (%): 85.52 -> 77.12Hurt
use low-rank restriction maps instead of diagonal (ablation)SheafHyperGNN · Congressaccuracy (%): 91.81 -> 74.83Hurt
use low-rank restriction maps instead of diagonal (ablation)SheafHyperGCN · Congressaccuracy (%): 90.37 -> 84.88Hurt
use general restriction maps instead of diagonal (ablation)SheafHyperGCN · Senateaccuracy (%): 66.33 -> 65.49No clear effect
use low-rank restriction maps instead of diagonal (ablation)SheafHyperGNN · Senateaccuracy (%): 68.73 -> 68.45No clear effect
increase stalk dimension d from 1 to >1 (ablation)SheafHyperGNN · synthetic heterophilic alpha=7accuracyImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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