AI Paper Atlas

Research map / Graph learning / GNN architectures

Neural Graph Generation from Graph Statistics

Kiarash Zahirnia, Yaochen Hu, Mark Coates, Oliver Schulte

NeurIPS 2023 · Graph learning › GNN architectures · approach: Theory of graph convolutional networks · 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
replace parametric statistics-based GGMs (BTER, Chung-Lu, Erdos-Renyi, SBM) with neural GenStatstatistics-based graph generation · 8 datasetsGNN-based MMD RBF: 12 out of 16 cases betterImproved
use GenStat instead of parametric baselines under Random/Pretrained-GNNstatistics-based graph generation · 8 datasetsGNN-based MMD RBFImproved
use GenStat instead of parametric baselines on Proteinstatistics-based graph generation · ProteinF1 PR: 83.72% vs. 71.08% (Random-GNN); 79.95% vs. 23.76% (Pretrained-GNN)Improved
use GenStat instead of parametric baselinesstatistics-based graph generation · most datasetsstatistics-based MMDs: 1-2 orders of magnitude betterImproved
train on graph statistics only (GenStat) vs on full adjacency matrices (deep GGMs)GenStat · real-world datasetsGNN-based MMD RBF: top two in 9 out of 10 scoresImproved
train on graph statistics only vs full adjacency matrices on synthetic regular graphs (Lobster, Grid, Triangle Grid)GenStat · synthetic datasetsGNN-based MMD RBF: Grid Random-GNN 0.53 vs BiGG 0.14Hurt
train under edge epsilon-LDP (epsilon 0.1 to 4) with GenStat on statistics vs BiGG on perturbed adjacencyGenStat · MUTAG, Lobster, GridRandom-GNN MMD RBFImproved
use GenStat generated graphs instead of BiGG for GNN benchmarking on link predictionGenStat · 8 datasetsbenchmark effectiveness (correlation): at least 0.5046 Pearson correlationMixed
train on statistics instead of adjacency matricesGenStat · training timetraining time: up to two orders of magnitude lower than fastest auto-regressive model; up to 5 times lower than GraphVAE-MMImproved
generate all-at-once (GenStat, GraphVAE-MM) instead of auto-regressivelyGenStat · generation timegeneration timeImproved

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