AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Graph Information Bottleneck for Subgraph Recognition

Junchi Yu, Tingyang Xu, Yu Rong, Yatao Bian, Junzhou Huang, Ran He

ICLR 2021 · Graph learning › Graph self-supervised learning · approach: Heterophily and imbalance on graphs · 3 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
Apply GIB to GINGIN · MUTAGaccuracy: 0.825$\pm$0.068 -> 0.839 $\pm$ 0.064No clear effect
Apply GIB to GINGIN · PROTEINSaccuracy: 0.707 $\pm$ 0.056 -> 0.749 $\pm$ 0.051No clear effect
Apply GIB to GINGIN · IMDB-BINARYaccuracy: 0.732 $\pm$ 0.048 -> 0.737 $\pm$ 0.070No clear effect
Apply GIB to GraphSAGEGraphSAGE · DDaccuracy: 0.729 $7$ 0.041 -> 0.781 $\pm$ 0.042Improved
Recognize an IB-subgraph with GIBGCN · MUTAG with 30% redundant edgesaccuracy: 0.667 -> 0.684Improved
Recognize an IB-subgraph with GIBGCN · MUTAG with 30% redundant edgesRecall: 0.324$\pm$ 0.049 -> 0.493$\pm$ 0.035Improved
Recognize an IB-subgraph with GIBGCN · MUTAG with 30% redundant edgesPrecision: 0.675$\pm$ 0.104 -> 0.692 $\pm$0.061No clear effect
Add mutual-information loss (ablation)GCN+GIB · ZINC QEDabsolute property bias: 0.43$\pm$ 0.15 -> 0.38$\pm$ 0.12Improved
Add mutual-information loss (ablation)GCN+GIB · ZINC DRD2absolute property bias: 0.21$\pm$ 0.13 -> 0.06$\pm$ 0.09Improved
Add mutual-information loss (ablation)GCN+GIB · ZINC HLM-CLintabsolute property bias: 0.48$\pm$ 0.34 -> 0.37$\pm$ 0.30Improved
Add mutual-information loss (ablation)GCN+GIB · ZINC MLM-CLintabsolute property bias: 1.20$\pm$ 0.97 -> 0.72$\pm$ 0.55Improved
Add connectivity loss (ablation)GCN+GIB · ZINC MLM-CLintabsolute property bias: 1.58$\pm$ 0.86 -> 0.72$\pm$ 0.55Improved

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