AI Paper Atlas

Research map / Graph learning / GNN architectures

Enhancing Graph Representation Learning with Localized Topological Features

Zuoyu Yan, Qi Zhao, Ze Ye, Tengfei Ma, Liangcai Gao, Zhi Tang, Yusu Wang, Chao Chen

JMLR 2025 · Graph learning › GNN architectures · approach: Graph structure learning · 0 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 pair-wise topological feature instead of node-wise concatenation (PEGN vs PEGN (node-wise)) (ablation)PEGN · Computers link predictionAUC-ROC: 97.0 -> 97.9Improved
add pair-wise topological feature instead of node classification features (PEGN vs PEGN (NC)) (ablation)PEGN · Cora link predictionAUC-ROC: 93.0 -> 94.9Improved
add pair-wise topological feature instead of node-wise concatenation (ablation)PEGN · Citeseer link predictionAUC-ROC: 92.9 -> 95.1Improved
use learnable filter function instead of Ollivier-Ricci curvature (PEGN (FF))PEGN · Cora node classificationaccuracy: 82.7 -> 81.8Hurt
use learnable filter function instead of Ollivier-Ricci curvature (PEGN (FF))PEGN · Computers node classificationaccuracy: 86.6 -> 86.2Hurt
combine persistence images (PEGN (PI)) instead of curvature filterPEGN · Photo node classificationaccuracy: 92.7 -> 90.2Hurt
use persistence image combination (PEGN (PI)) in link predictionPEGN · Computers link predictionAUC-ROC: 97.9 -> 98.3Improved
use learnable filter function (PEGN (FF)) in link predictionPEGN · PPI link prediction (graph 1)AUC-ROC: 83.71 -> 85.47Improved
use PI+ representation instead of original persistence image with GCN (ablation)GCN · Photo node classificationaccuracy: 92.5 -> 93.0Improved
use PI+ representation instead of original persistence image with GCN (ablation)GCN · CS node classificationaccuracy: 92.3 -> 93.0Improved
add persistence-image topological features to GCN (GCN(+PI+) vs GCN) (ablation)GCN · Computers node classificationaccuracy: 82.6 -> 86.9Improved
swap in the relaxed shortest-path filter used in the proof (PEGN (new)) (ablation)PEGN · Physics node classificationaccuracy: 94.3 -> 94.4No clear effect

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