AI Paper Atlas

Research map / Graph learning / GNN architectures

Nested Graph Neural Networks

NeurIPS 2021 · Graph learning › GNN architectures · approach: Subgraph and distance-aware GNNs · 32 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Replace rooted-subtree node representations with pooled rooted-subgraph representationsGIN · EXPtest accuracy: 50.0±0.00 -> 99.9±0.26Helped
Use nested rooted-subgraph processingGIN · ogbg-molpcba testAP: 27.03{ ±0.23} -> 28.32{ ±0.41}Helped
Use checkpoint ensembleNested GIN* · ogbg-molpcba testAP: 28.32{ ±0.41} -> 30.07{ ±0.37}Helped
Use nested rooted-subgraph processingGIN · ENZYMESaccuracy: 38.3 ±6.4 -> 29.0 ±8.0Hurt
Add Distance Encoding (ablation)Nested 1-GNN · QM9 μMAE: 0.466 -> 0.428Helped
Add Distance Encoding (ablation)Nested 1-GNN · QM9 ε_HOMOMAE: 0.00292 -> 0.00265Helped
Add Distance Encoding (ablation)Nested 1-GNN · QM9 C_vMAE: 0.111 -> 0.174Hurt
Add Distance Encoding (ablation)Nested 1-2-GNN · QM9 U_0MAE: 0.349 -> 0.252Hurt

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