Research map / Label-efficient and robust learning / Recognition and long-tail
Robust Negative Sampling for Network Embedding
Mohammadreza Armandpour, Patrick Ding, Jianhua Huang, Xia Hu
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 |
|---|---|---|---|
| replace NS with R-NS (distant negative sampling plus norm penalty) | LINE · BlogCatalog, Type II, 50% labeled | Micro-F1: 0.12 -> 0.25 | Improved |
| replace NS with R-NS | LINE · Flickr, Type II, 50% labeled | Micro-F1: 0.46 -> 0.57 | Improved |
| replace NS with R-NS | LINE · Wikipedia, Type I, 50% labeled | Micro-F1: 0.47 -> 0.48 | Improved |
| add the norm penalty to R-NS (lambda 0 -> formula) (ablation) | R-NS · BlogCatalog, Type II | Micro-F1: 0.15 -> 0.25 | Improved |
| add the norm penalty to R-NS (ablation) | R-NS · Wikipedia, Type I | Micro-F1: 0.42 -> 0.48 | Improved |
| exclude neighbors from negative sampling without penalty (ablation) | NS · BlogCatalog, Type I | Micro-F1: 0.36 -> 0.34 | Hurt |
| use R-NS penalty formula instead of NN-NS implicit degree-based penalty (ablation) | R-NS · Flickr, Type II | Micro-F1: 0.41 -> 0.57 | Improved |
| vary lambda away from formula value (ablation) | R-NS · BlogCatalog and Wikipedia | Micro-F1 | Mixed |
Closest papers by meaning
- Reconciling Competing Sampling Strategies of Network Embedding (NeurIPS 2023 · similarity 0.89)
- Asymptotics of Network Embeddings Learned via Subsampling (JMLR 2023 · similarity 0.88)
- Comprehensive Analysis of Negative Sampling in Knowledge Graph Representation Learning (ICML 2022 · similarity 0.87)
- RSDNE: Exploring Relaxed Similarity and Dissimilarity from Completely-Imbalanced Labels for Network Embedding (AAAI 2018 · similarity 0.87)
- LoNe Sampler: Graph Node Embeddings by Coordinated Local Neighborhood Sampling (AAAI 2023 · similarity 0.86)
- Does GCL Need a Large Number of Negative Samples? Enhancing Graph Contrastive Learning with Effective and Efficient Negative Sampling (AAAI 2025 · similarity 0.85)
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.
