Research map / Graph learning / Graph self-supervised learning
Node Representation Learning in Graph via Node-to-Neighbourhood Mutual Information Maximization
Check 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 |
|---|---|---|---|
| maximize mutual information between node and neighbourhood representations | N2N · node classification across six datasets | micro-F1 | Helped |
| choose positives using TAPS structural dependency instead of random neighbour sampling (ablation) | N2N-TAPS-1 (JL) · Cora | micro-F1: 83.46$\pm$0.18 -> 85.46$\pm$0.08 | Helped |
| choose positives using TAPS structural dependency instead of random neighbour sampling (ablation) | N2N-TAPS-1 (JL) · Coauthor Physics | micro-F1: 88.75$\pm$0.12 -> 90.70$\pm$0.20 | Helped |
| increase TAPS positive count from one to five (ablation) | N2N-TAPS (JL) · Cora | micro-F1: 85.46$\pm$0.08 -> 87.10$\pm$0.08 | Helped |
| increase TAPS positive count from one to five (ablation) | N2N-TAPS (JL) · Coauthor Physics | micro-F1: 90.70$\pm$0.20 -> 91.38$\pm$0.12 | Helped |
| use one TAPS-selected positive rather than aggregating all neighbours (ablation) | N2N (JL) · node classification across datasets | micro-F1 | Mixed |
| increase randomly sampled positive count from one to five (ablation) | N2N-Random (JL) · Amazon Photo and Coauthor Physics | micro-F1 | Mixed |
| replace random neighbour sampling with TAPS (ablation) | GraphSAGE-Mean · node classification datasets | micro-F1 | Helped |
| use more TAPS-selected neighbours (ablation) | GraphSAGE-Mean-TAPS · node classification datasets | micro-F1 | Helped |
Closest papers by meaning
- LoNe Sampler: Graph Node Embeddings by Coordinated Local Neighborhood Sampling (AAAI 2023 · similarity 0.90)
- Robust Node Classification on Graph Data with Graph and Label Noise (AAAI 2024 · similarity 0.88)
- Node Identifiers: Compact, Discrete Representations for Efficient Graph Learning (ICLR 2025 · similarity 0.87)
- Sterling: Synergistic Representation Learning on Bipartite Graphs (AAAI 2024 · similarity 0.87)
- Learning Graph Representation via Graph Entropy Maximization (ICML 2024 · similarity 0.87)
- Learning Representations for Hierarchies with Minimal Support (NeurIPS 2024 · similarity 0.87)
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.