Research map / Graph learning / Graph self-supervised learning
Directed Graph Contrastive Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 curriculum learning over views (easy-to-difficult) vs fixed views (ablation) | DiGCL · CITESEER | accuracy: 66.99 -> 67.42 | Helped |
| add curriculum learning over views vs fixed views (ablation) | DiGCL · CORA-ML | accuracy: 75.86 -> 77.53 | Helped |
| easy-to-difficult order vs anti-curriculum order (ablation) | DiGCL · CORA-ML | accuracy: 76.12 -> 77.53 | Helped |
| easy-to-difficult order vs random order (ablation) | DiGCL · PUBMED | accuracy: 80.75 -> 80.69 | No clear effect |
| easy-to-difficult order vs random order (ablation) | DiGCL · AM-PHOTO | accuracy: 89.03 -> 89.41 | Helped |
| use Laplacian perturbation instead of random edge removal (ablation) | DiGCL · CITESEER | accuracy: 64.97 -> 67.42 | Helped |
| use Laplacian perturbation instead of random edge removal (ablation) | DiGCL · PUBMED | accuracy: 79.66 -> 80.69 | No clear effect |
| use Laplacian perturbation instead of random edge removal (ablation) | DiGCL · AM-PHOTO | accuracy: 88.45 -> 89.41 | Helped |
| choose log pacing function over other pacing functions (ablation) | DiGCL · node classification | accuracy | Helped |
| plug Laplacian perturbation into MVGRL and GCA (ablation) | MVGRL, GCA · directed graph node classification | accuracy | Helped |
| plug DiGCL into MVGRL for graph classification (ablation) | MVGRL · MUTAG, PTC, IMDB-BIN | accuracy | No clear effect |
Closest papers by meaning
- Revisiting Graph Contrastive Learning from the Perspective of Graph Spectrum (NeurIPS 2022 · similarity 0.92)
- An Empirical Study of Graph Contrastive Learning (NeurIPS 2021 · similarity 0.92)
- Perfect Alignment May be Poisonous to Graph Contrastive Learning (ICML 2024 · similarity 0.91)
- ReGCL: Rethinking Message Passing in Graph Contrastive Learning (AAAI 2024 · similarity 0.91)
- Community-Invariant Graph Contrastive Learning (ICML 2024 · similarity 0.90)
- A New Mechanism for Eliminating Implicit Conflict in Graph Contrastive Learning (AAAI 2024 · similarity 0.90)
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.