Research map / Label-efficient and robust learning / Semi-supervised learning
SNEQ: Semi-Supervised Attributed Network Embedding with Attention-Based Quantisation
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 SNEQ continuous embeddings instead of Graph2Gauss | SNEQ · Link prediction cDBLP | AUC: 88.18 -> 92.12 | Helped |
| use SNEQ instead of Graph2Gauss | SNEQ · Link prediction Cora ML | AUC: 97.63 -> 97.21 | Hurt |
| self-attention quantisation (SNEQ#) instead of hashing (BANE) | SNEQ# · Link prediction DBLP | AUC: 90.63 -> 93.60 | Helped |
| quantisation instead of NetHash for node recommendation | SNEQ# · Node recommendation Cora ML | NDCG@50: 37.54 -> 41.82 | Helped |
| quantise embeddings (SNEQ#) versus float embeddings | SNEQ · cDBLP storage | storage (MB): 155.31 -> 2.697 | Helped |
| quantised codes versus float embeddings for retrieval | SNEQ · cDBLP retrieval | time (ms): 3,923.5 -> 59.6 | Helped |
| use M=8, K=16 instead of M=16, K=8 (ablation) | SNEQ# · Link prediction cDBLP | AUC: 91.24 -> 90.79 | Hurt |
| use M=8, K=16 instead of M=16, K=8 (ablation) | SNEQ# · Node recommendation DBLP | NDCG@50: 9.79 -> 9.03 | Hurt |
| use M=8, K=16 instead of M=16, K=8 (ablation) | SNEQ# · cDBLP retrieval | time (ms): 59.6 -> 26.1 | Helped |
Closest papers by meaning
- Learning Signed Network Embedding via Graph Attention (AAAI 2020 · similarity 0.87)
- Semi-supervisedly Co-embedding Attributed Networks (NeurIPS 2019 · similarity 0.87)
- RSDNE: Exploring Relaxed Similarity and Dissimilarity from Completely-Imbalanced Labels for Network Embedding (AAAI 2018 · similarity 0.87)
- Unsupervised Attributed Multiplex Network Embedding (AAAI 2020 · similarity 0.87)
- How Low Can You Go? Searching for the Intrinsic Dimensionality of Complex Networks using Metric Node Embeddings (ICLR 2025 · similarity 0.87)
- PRUNE: Preserving Proximity and Global Ranking for Network Embedding (NeurIPS 2017 · 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.