AI Paper Atlas

Research map / Graph learning / GNN architectures

A Variational Edge Partition Model for Supervised Graph Representation Learning

Yilin He, Chaojie Wang, Hao Zhang, Bo Chen, Mingyuan Zhou

NeurIPS 2022 · Graph learning › GNN architectures · approach: Graph structure learning · 1 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
Learn task-relevant overlapping communities and partition edges for community-specific GNN aggregationVEPM · Coraaccuracy: 83.7 -> 84.3Improved
Learn task-relevant overlapping communities and partition edges for community-specific GNN aggregationVEPM · Pubmedaccuracy: 80.5 -> 82.4Improved
Learn task-relevant overlapping communities and partition edges for community-specific GNN aggregationVEPM · MUTAGaccuracy: 92.6 ± 4.2 -> 93.6 ± 3.4Improved
Learn task-relevant overlapping communities and partition edges for community-specific GNN aggregationVEPM · NCI1accuracy: 86.0 ± 1.8 -> 83.9 ± 1.8Hurt
Replace the learned edge partition with an even partition (ablation)VEPM · IMDB-Baccuracy: 76.7 ± 3.1 -> 74.2 ± 4.3Hurt
Replace the learned edge partition with a fixed random partition (ablation)VEPM · IMDB-Baccuracy: 76.7 ± 3.1 -> 60.5 ± 4.7Hurt
Replace the GNN representation composer with an FC layer (ablation)VEPM with meaningful partitions · PTCaccuracy: 75.6 ± 5.9 -> 68.7 ± 4.3Hurt
Increase edge-partition temperature on random partitions (ablation)VEPM · MUTAGaccuracy: 79.2 ± 7.9 -> 89.8 ± 4.6Improved
Use subgraph-sampling-based scalable training (ablation)VEPM · Coraaccuracy: 84.3 ± 0.1 -> 83.4 ± 0.1Hurt
Use subgraph-sampling-based scalable training (ablation)VEPM · Coraepoch duration: 0.19s -> 0.02sImproved
Add community affiliation scores to hand-crafted input features (ablation)VEPM · IMDB-Baccuracy: 80.3 ± 2.0 -> 76.7 ± 3.1No clear effect
Add unsupervised pretraining before supervised finetuning (ablation)VEPM · IMDB-Baccuracy: 61.0 ± 3.6 -> 76.7 ± 3.1Improved

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