AI Paper Atlas

Research map / Graph learning / GNN architectures

Distance-Restricted Folklore Weisfeiler-Leman GNNs with Provable Cycle Counting Power

Junru Zhou, Jiarui Feng, Xiyuan Wang, Muhan Zhang

NeurIPS 2023 · Graph learning › GNN architectures · approach: Subgraph and distance-aware GNNs · 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
increase restriction distance d from 2 to 3 (ablation)d-DRFWL(2) GNN · Synthetic 7-cycle countingnormalized MAE: 0.0362 -> 0.0176Improved
increase d from 2 to 3 (ablation)d-DRFWL(2) GNN · Synthetic 6-cycle countingnormalized MAE: 0.0087 -> 0.0099No clear effect
increase d from 2 to 3 (ablation)d-DRFWL(2) GNN · Synthetic 4-cycle countingnormalized MAE: 0.0015 -> 0.0020No clear effect
reduce d from 2 to 1 (ablation)d-DRFWL(2) GNN · Synthetic cycle countingnormalized MAEHurt
use 2-DRFWL(2) GNN instead of NGNN (subgraph GNN)2-DRFWL(2) GNN · QM9 efficiencyGPU memory (GB): 13.72 -> 2.31Improved
use 2-DRFWL(2) GNN instead of I2-GNN2-DRFWL(2) GNN · QM9 efficiencypreprocessing time (s): 5287 -> 430Improved
use 2-DRFWL(2) GNN instead of I2-GNN2-DRFWL(2) GNN · QM9 efficiencytraining time (s/epoch): 209.9 -> 141.9Improved
use 2-DRFWL(2) GNN instead of NGNN2-DRFWL(2) GNN · QM9 efficiencytraining time (s/epoch): 107.8 -> 141.9No clear effect
use 2-DRFWL(2) GNN instead of I2-GNN2-DRFWL(2) GNN · ogbg-molhiv efficiencyGPU memory (GB): 11.07 -> 4.44Improved
use 2-DRFWL(2) GNN instead of I2-GNN2-DRFWL(2) GNN · QM9 U_0MAE: 0.211 -> 0.156Improved
use 2-DRFWL(2) GNN instead of I2-GNN2-DRFWL(2) GNN · QM9 ZPVEMAE: 0.00014 -> 0.00017Hurt
use 2-DRFWL(2) GNN instead of I2-GNN2-DRFWL(2) GNN · QM9 muMAE: 0.428 -> 0.346Improved

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