AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Automated Data Augmentations for Graph Classification

Youzhi Luo, Michael Curtis McThrow, Wing Yee Au, Tao Komikado, Kanji Uchino, Koji Maruhashi, Shuiwang Ji

ICLR 2023 · Graph learning › Graph self-supervised learning · approach: Augmentation-free graph learning · 9 citations (OpenAlex)

Read the paper

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 learned per-element transformations with uniform transformations (remove learnable graph transformation) (ablation)GraphAug · PROTEINS, IMDB-BINARY, NCI1accuracy: PROTEINS 0.722 vs 0.696; IMDB-BINARY 0.762 vs 0.724; NCI1 0.816 vs 0.760Improved
Random augmentation category selection instead of learned selection (ablation)GraphAug · PROTEINS, IMDB-BINARY, NCI1accuracy: -0.020 PROTEINS, -0.016 IMDB-BINARY, -0.020 NCI1Hurt
Single augmentation step (T=1) instead of sequential multi-step augmentation (ablation)GraphAug · PROTEINS, IMDB-BINARY, NCI1accuracy: -0.010 PROTEINS, -0.009 IMDB-BINARY, -0.007 NCI1Hurt
Uniform MaskNF augmentation (label-violating transformation)GIN classifier · COLORStesting accuracy: 0.507 vs 0.578 without augmentationHurt
Uniform PerturbEdge augmentationGIN classifier · TRIANGLEStesting accuracy: 0.303 vs 0.506 without augmentationHurt
Learned label-invariant augmentation (GraphAug) instead of no augmentationGIN classifier · COLORS and TRIANGLEStesting accuracy: +0.055 COLORS; +0.007 TRIANGLESImproved
GraphAug augmentation vs no augmentation on graph benchmarksGIN classifier · PROTEINS, IMDB-BINARY, COLLAB, MUTAG, NCI109, NCI1, ogbg-molhivaccuracy / ROC-AUC: gains from +0.002 (ogbg-molhiv 0.756→0.774 is +0.018) to +0.026 (MUTAG); largest gain IMDB +0.031Improved
Uniform augmentations on molecular datasets (MaskNF, DropNode, PerturbEdge)GIN classifier · MUTAG, NCI109, NCI1accuracy: e.g., NCI109 PerturbEdge 0.555 vs 0.794; NCI1 PerturbEdge 0.545 vs 0.804Hurt
Uniform DropNode augmentation on ogbg-molhivGIN classifier · ogbg-molhivROC-AUC: 0.717 vs 0.756Hurt
Uniform MaskNF augmentation on COLLABGIN classifier · COLLABaccuracy: +0.009 vs 0.806Improved

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