AI Paper Atlas

Research map / Graph learning / GNN architectures

High-Order Pooling for Graph Neural Networks with Tensor Decomposition

Chenqing Hua, Guillaume Rabusseau, Jian Tang

NeurIPS 2022 · Graph learning › GNN architectures · approach: Simple and scalable GNNs · 6 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
add high-order nonlinear CP pooling alongside low-order linear sum interactions (ablation)tGNN · Coraaccuracy: 0.8623$\pm$0.0107 -> 0.8780$\pm$0.0158Improved
add high-order nonlinear CP pooling alongside low-order linear sum interactions (ablation)tGNN · Pubmedaccuracy: 0.8531$\pm$0.0009 -> 0.9018$\pm$0.0015Improved
add high-order nonlinear CP pooling alongside low-order linear sum interactions (ablation)tGNN · ZINCMAE: 0.440$\pm$0.010 -> 0.301$\pm$0.008Improved
add low-order linear sum interactions to nonlinear CP pooling (ablation)tGNN · Coraaccuracy: 0.8655$\pm$0.0375 -> 0.8780$\pm$0.0158Improved
add low-order linear sum interactions to nonlinear CP pooling (ablation)tGNN · Pubmedaccuracy: 0.8679$\pm$0.0103 -> 0.9018$\pm$0.0015Improved
add low-order linear sum interactions to nonlinear CP pooling (ablation)tGNN · ZINCMAE: 0.407$\pm$0.025 -> 0.301$\pm$0.008Improved
replace standard GNN pooling with high-order CP pooling and its linear interaction channeltGNN · Coraaccuracy: 0.8778$\pm$0.0096 -> 0.8808$\pm$0.0131Improved
replace standard GNN pooling with high-order CP pooling and its linear interaction channeltGNN · Pubmedaccuracy: 0.8890$\pm$0.0032 -> 0.9080$\pm$0.0018Improved
replace standard GNN pooling with high-order CP pooling and its linear interaction channeltGNN · PROTEINSAUC: 0.8580$\pm$0.0017 -> 0.8255$\pm$0.0049Hurt
increase CP decomposition rank (ablation)tGNN · Coraaccuracy: 85.55 -> 87.62Improved
increase sampled neighbors from 3 to 10 (ablation)tGNN · Coraaccuracy: 85.55 -> 85.40; 86.25 -> 86.35; 86.06 -> 86.83; 86.76 -> 86.77; 86.97 -> 87.42; 87.33 -> 87.22; 87.62 -> 87.87Mixed

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