Research map / Graph learning / Graph self-supervised learning
Feature-Centric Unsupervised Node Representation Learning Without Homophily Assumption
Sunwoo Kim, Soo Yong Lee, Kyungho Kim, Hyunjin Hwang, Jaemin Yoo, Kijung Shin
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 clustering-based Step 1 training (ablation) | FUEL · 14 datasets | classification average ranking: 8.3 -> 2.4 | Improved |
| add refinement Step 2 (ablation) | FUEL · 14 datasets | classification average ranking: 8.6 -> 2.4 | Improved |
| add skip connection in the refiner (ablation) | FUEL · 14 datasets | classification average ranking: 5.5 -> 2.4 | Improved |
| add exponential function in the loss (ablation) | FUEL · 14 datasets | classification average ranking: 6.1 -> 2.4 | Improved |
| add clustering-based Step 1 training (ablation) | FUEL · 14 datasets | clustering average ranking: 9.5 -> 2.4 | Improved |
| add refinement Step 2 (ablation) | FUEL · 14 datasets | clustering average ranking: 9.3 -> 2.4 | Improved |
| add skip connection in the refiner (ablation) | FUEL · 14 datasets | clustering average ranking: 4.9 -> 2.4 | Improved |
| add exponential function in the loss (ablation) | FUEL · 14 datasets | clustering average ranking: 6.4 -> 2.4 | Improved |
| add Step 1 clustering scheme (ablation) | FUEL · Texas | accuracy: 63.2 -> 84.6 | Improved |
| add Step 1 clustering scheme (ablation) | FUEL · Squirrel | accuracy: 46.6 -> 65.2 | Improved |
| add refinement Step 2 (ablation) | FUEL · Cora | accuracy: 78.9 -> 83.8 | Improved |
| add skip connection in the refiner (ablation) | FUEL · Penn94 | accuracy: 71.6 -> 74.6 | Improved |
Closest papers by meaning
- Beyond Smoothing: Unsupervised Graph Representation Learning with Edge Heterophily Discriminating (AAAI 2023 · similarity 0.89)
- PINE: Universal Deep Embedding for Graph Nodes via Partial Permutation Invariant Set Functions (TPAMI 2021 · similarity 0.89)
- Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking (ICLR 2018 · similarity 0.88)
- Self-Supervised Node Representation Learning via Node-to-Neighbourhood Alignment (TPAMI 2024 · similarity 0.88)
- Rethinking Kernel Methods for Node Representation Learning on Graphs (NeurIPS 2019 · similarity 0.88)
- Co-Embedding of Nodes and Edges With Graph Neural Networks (TPAMI 2020 · similarity 0.88)
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.
