AI Paper Atlas

Research map / Graph learning / GNN architectures

On the Robustness of Graph Neural Diffusion to Topology Perturbations

Yang Song, QIYU KANG, Sijie Wang, Zhao Kai, Wee Peng Tay

NeurIPS 2022 · Graph learning › GNN architectures · approach: Physics-inspired GNNs · 12 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
Use Beltrami diffusion rather than standard non-PDE GNNsNode classification models · PubMed under SPEIT and TDGIAaccuracy (%): 84.36 $\pm$ 0.58 -> 86.66 $\pm$ 0.68; 84.00 $\pm$ 1.12 -> 85.56 $\pm$ 0.91Improved
Use Beltrami diffusion rather than GNNGuardNode classification models · Cora under SPEIT and TDGIAaccuracy (%): 78.50 $\pm$ 2.27 vs 61.87 $\pm$ 0.49; 78.92 $\pm$ 1.80 vs 62.84 $\pm$ 1.17Hurt
Combine Beltrami with GNNGuardBeltrami · Citeseer and PubMed under SPEIT and TDGIAaccuracy (%): Citeseer: 66.46 $\pm$ 1.33 -> 69.90 $\pm$ 0.44; PubMed SPEIT: 86.66 $\pm$ 0.68 -> 87.68 $\pm$ 0.03Improved
Replace Beltrami flow with heat flow (ablation)Graph neural PDE · Cora and Citeseer under SPEITaccuracy (%): Cora: 61.87 $\pm$ 0.49 -> 38.66 $\pm$ 2.40; Citeseer: 66.46 $\pm$ 1.33 -> 22.95 $\pm$ 5.07Hurt
Replace heat flow with mean curvature or Beltrami flow (ablation)Graph neural PDE · Cora and Citeseer under SPEITaccuracy (%): Cora heat 38.66 $\pm$ 2.40; mean curvature 60.67 $\pm$ 1.31; Beltrami 61.87 $\pm$ 0.49. Citeseer heat 22.95 $\pm$ 5.07; Improved
Replace Beltrami flow with mean curvature flow (ablation)Graph neural PDE · Citeseer TDGIAaccuracy (%): 65.77 $\pm$ 1.28 -> 66.83 $\pm$ 1.59Improved
Replace Beltrami flow with GRAND/BLEND (ablation)Graph neural PDE · Cora SPEIT and PubMed SPEITaccuracy (%): Cora: 61.87 $\pm$ 0.49 -> 55.67 $\pm$ 3.60; PubMed: 86.66 $\pm$ 0.68 -> 39.39 $\pm$ 0.26Hurt
Remove spectral normalization/Lipschitz constraint (ablation)Beltrami · Cora SPEIT, 4 layersaccuracy (%): 61.87 $\pm$ 0.49 -> 56.79 $\pm$ 1.52Hurt
Reduce the model to 1, 2, or 3 layers rather than 4 (ablation)Beltrami · Cora SPEITaccuracy (%): 4 layers: 61.87 $\pm$ 0.49; 1 layer: 59.10 $\pm$ 3.20; 2 layers: 63.58 $\pm$ 1.39; 3 layers: 60.52 $\pm$ 1.56Mixed
Use Dopri5 instead of implicit Adam with τ=2 (ablation)Beltrami solver · Citeseer SPEITaccuracy (%): 66.46 $\pm$ 1.33 -> 66.96 $\pm$ 1.46Improved
Increase injection attack strength (ablation)Beltrami · Cora, Citeseer, and PubMedaccuracy (%)No clear effect
Make heat diffusivity a trainable shared edge weight (ablation)Heat flow · Citeseer SPEITaccuracy (%): Heat+ 54.61 $\pm$ 2.92; Heat 22.95 $\pm$ 5.07Improved

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