Research map / Learning theory and algorithms / Prediction and decision losses
Conditional Network Embeddings
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 |
|---|---|---|---|
| use a degree prior instead of a uniform prior (ablation) | CNE · Facebook | AUC: 0.9905 -> 0.9909 | Helped |
| use a degree prior instead of a uniform prior (ablation) | CNE · PPI | AUC: 0.8908 -> 0.9115 | Helped |
| use a degree prior instead of a uniform prior (ablation) | CNE · arXiv | AUC: 0.9865 -> 0.9882 | Helped |
| use a degree prior instead of a uniform prior (ablation) | CNE · BlogCatalog | AUC: 0.9190 -> 0.9636 | Helped |
| use a degree prior instead of a uniform prior (ablation) | CNE · Wikipedia | AUC: 0.8417 -> 0.9158 | Helped |
| use a degree prior instead of a uniform prior (ablation) | CNE · StudentDB | AUC: 0.9300 -> 0.9439 | Helped |
| add a block-structure prior to the degree prior (ablation) | CNE · StudentDB | AUC: from $94.39\%$ to $98.30\%$ | Helped |
| predict labels as links using CNE-LP rather than logistic regression on the embedding (ablation) | CNE multi-label classification · BlogCatalog | Macro-F1: 0.1833 -> 0.2935 | Helped |
| predict labels as links using CNE-LP rather than logistic regression on the embedding (ablation) | CNE multi-label classification · PPI | Macro-F1: 0.1484 -> 0.2639 | Helped |
| predict labels as links using CNE-LP rather than logistic regression on the embedding (ablation) | CNE multi-label classification · Wikipedia | Macro-F1: 0.1370 -> 0.3374 | Helped |
| use CNE-LP (block+degree) instead of LINE | multi-label classification · Wikipedia | Micro-F1: 0.5193 -> 0.4839 | Hurt |
Closest papers by meaning
- RSDNE: Exploring Relaxed Similarity and Dissimilarity from Completely-Imbalanced Labels for Network Embedding (AAAI 2018 · similarity 0.90)
- Reconciling Competing Sampling Strategies of Network Embedding (NeurIPS 2023 · similarity 0.90)
- How Low Can You Go? Searching for the Intrinsic Dimensionality of Complex Networks using Metric Node Embeddings (ICLR 2025 · similarity 0.88)
- Conditional Similarity Networks (CVPR 2017 · similarity 0.88)
- Structural Deep Embedding for Hyper-Networks (AAAI 2018 · similarity 0.88)
- Link Prediction via Subgraph Embedding-Based Convex Matrix Completion (AAAI 2018 · 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.