AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Let Invariant Rationale Discovery Inspire Graph Contrastive Learning

Sihang Li, Xiang Wang, An Zhang, Yingxin Wu, Xiangnan He, Tat-Seng Chua

ICML 2022 · Graph learning › Graph self-supervised learning · approach: Graph contrastive learning analysis · 23 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
Use rationale-aware views instead of random node-dropping views (ablation)RGCL on MoleculeNet transfer tasks · MoleculeNet transfer, AVG.ROC-AUC (%): 70.83 -> 73.16Improved
Include the independence loss term (ablation)RGCL on MoleculeNet transfer tasks · MoleculeNet transfer, AVG.ROC-AUC (%): 72.60 -> 73.16Improved
Use rationale-aware views instead of GraphCL's random augmentationsGraphCL on MoleculeNet transfer tasks · MoleculeNet transfer, AVG.GAIN: 3.77 to 6.16Improved
Use rationale-aware views instead of random augmentationGraphCL on MNIST-Superpixel · MNIST-Superpixel, 10% MNISTaccuracy (%): 93.11 -> 94.92Improved
Use rationale-aware views instead of random augmentationGraphCL on MNIST-Superpixel · MNIST-Superpixel, 1% MNISTaccuracy (%): 83.41 -> 83.84Improved
Use RGCL rather than GraphCLMoleculeNet transfer tasks · HIVROC-AUC (%): 78.47 -> 77.90Hurt
Remove independence loss term (ablation)RGCL on MoleculeNet transfer tasks · MoleculeNet transfer, AVG.ROC-AUC (%): 73.16 -> 72.60Hurt
Replace rationale-aware augmentation with random node discarding (ablation)RGCL on MoleculeNet transfer tasks · MoleculeNet transfer, AVG.ROC-AUC (%): 73.16 -> 70.83Hurt
Increase the contrastive temperature above 0.2 (ablation)RGCL · MoleculeNet and TU datasetsaverage performanceHurt
Set contrastive temperature too small (e.g., 0.05) (ablation)RGCL · MoleculeNet and TU datasetsaverage performanceHurt
Overemphasize the independence regularizer with a large lambda (ablation)RGCL · MoleculeNet and TU datasetsaverage performanceHurt
Use a proper independence regularizer (lambda=0.1) (ablation)RGCL · MoleculeNet and TU datasetsaverage performanceImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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