Research map / Graph learning / Graph self-supervised learning
Group Contrastive Self-Supervised Learning on Graphs
Xinyi Xu, Cheng Dan Deng, Yaochen Xie, Shuiwang Ji
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add group contrast to GraphCL (GroupCL, non-parameterized CLUB) (ablation) | GraphCL · NCI1 | accuracy: 77.87 -> 81.69 | Improved |
| add group contrast to GraphCL (GroupCL, non-parameterized CLUB) (ablation) | GraphCL · MUTAG | accuracy: 86.60 -> 91.67 | Improved |
| add group contrast to GraphCL (GroupCL, non-parameterized CLUB) (ablation) | GraphCL · COLLAB | accuracy: 71.36 -> 76.38 | Improved |
| add group contrast to GraphCL (GroupCL, non-parameterized CLUB) (ablation) | GraphCL · IMDB-B | accuracy: 71.14 -> 73.52 | Improved |
| use parameterized CLUB estimator instead of none (ablation) | GroupCL · NCI1 | accuracy: 77.87 -> 79.87 | Improved |
| use non-parameterized instead of parameterized CLUB estimator (ablation) | GroupCL · NCI1 | accuracy: 79.87 -> 81.69 | Improved |
| use non-parameterized instead of parameterized CLUB estimator (ablation) | GroupCL · MUTAG | accuracy: 90.63 -> 91.67 | Improved |
| use non-parameterized instead of parameterized CLUB estimator (ablation) | GroupCL · IMDB-B | accuracy: 73.38 -> 73.52 | Improved |
| increase number of groups p from 1 to 4 (ablation) | GroupCL · DD | accuracy: 73.92 -> 76.76 | Improved |
| increase number of groups p from 1 to 4 (ablation) | GroupCL · IMDB-B | accuracy: 71.28 -> 73.52 | Improved |
| increase number of groups p from 4 to 5 (ablation) | GroupCL · IMDB-B | accuracy: 73.52 -> 70.64 | Hurt |
| increase number of groups p from 1 to 4 (ablation) | GroupCL · MUTAG | accuracy: 90.85 -> 91.67 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- InfoGraph — InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization (ICLR 2020) · better on 8 of 8 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
- Rethinking and Scaling Up Graph Contrastive Learning: An Extremely Efficient Approach with Group Discrimination (NeurIPS 2022 · similarity 0.92)
- S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning (ICML 2024 · similarity 0.91)
- Exploitation of a Latent Mechanism in Graph Contrastive Learning: Representation Scattering (NeurIPS 2024 · similarity 0.89)
- Beyond Homophily: Graph Contrastive Learning with Macro-Micro Message Passing (AAAI 2025 · similarity 0.89)
- Does GCL Need a Large Number of Negative Samples? Enhancing Graph Contrastive Learning with Effective and Efficient Negative Sampling (AAAI 2025 · similarity 0.89)
- Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning (AAAI 2024 · similarity 0.89)
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.
