Research map / Probabilistic and scientific modeling / Molecules and proteins
Adaptive Shrinkage Estimation for Streaming Graphs
Nesreen Ahmed, Nick Duffield
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 James-Stein shrinkage to APS | APS · soc-flickr, soc-livejournal, soc-youtube-snap, wiki-Talk, web-BerkStan-dir, cit-Patents, soc-orkut-dir | MSE: 22.30K -> 295.13; 214.80 -> 16.11; 11.35 -> 6.68; 7.70 -> 5.32; 7.32K -> 561.20; 6.02 -> 3.03; 2.08K -> 70.79 | Improved |
| Use APS JS instead of Uniform | Triangle-weighted streaming graph estimation · soc-flickr | MSE: 6.3K -> 295.13 | Improved |
| Use APS JS instead of Uniform | Triangle-weighted streaming graph estimation · soc-livejournal | relative spectral norm: 0.429 -> 0.0089 | Improved |
| Use APS JS instead of Triest | Triangle-weighted streaming graph estimation · soc-orkut-dir | MSE: 613.89 -> 70.79 | Improved |
| Use APS instead of Uniform | Triangle-weighted streaming graph estimation · soc-orkut-dir | MSE: 467.90 -> 2.08K | Hurt |
| Use APS instead of Uniform or Triest | Triangle-weighted streaming graph estimation · All seven reported graphs | relative spectral norm | Improved |
| Replace adaptive weights with fixed arrival-time weights (ablation) | APS without shrinkage · soc-flickr | MSE: 22.30K -> 4907.21 | Improved |
| Replace adaptive weights with fixed arrival-time weights (ablation) | APS without shrinkage · soc-livejournal | MSE: 214.80 -> 94.46 | Improved |
| Replace adaptive weights with fixed arrival-time weights (ablation) | APS without shrinkage · soc-youtube-snap and wiki-Talk | MSE: 11.35 -> 24.78; 7.70 -> 78.69 | Hurt |
| Replace adaptive weights with fixed arrival-time weights (ablation) | APS without shrinkage · web-BerkStan-dir and cit-Patents | MSE: 7.32K -> 1723.63; 6.02 -> 6.45 | Improved |
| Replace adaptive weights with fixed arrival-time weights (ablation) | APS without shrinkage · soc-orkut-dir | MSE: 2.08K -> 405.86 | Improved |
| Add James-Stein shrinkage (ablation) | Non-adaptive priority sampling · soc-flickr, soc-livejournal, web-BerkStan-dir, cit-Patents | MSE: 4907.21 -> 2174.9; 94.46 -> 69.97; 1723.63 -> 1236.3; 6.45 -> 5.67 | Improved |
Closest papers by meaning
- Faster Streaming and Scalable Algorithms for Finding Directed Dense Subgraphs in Large Graphs (ICML 2024 · similarity 0.86)
- Efficient Streaming Algorithms for Graphlet Sampling (NeurIPS 2024 · similarity 0.86)
- Learning-Augmented Streaming Algorithms for Correlation Clustering (NeurIPS 2025 · similarity 0.85)
- New Parallel and Streaming Algorithms for Directed Densest Subgraph (NeurIPS 2025 · similarity 0.85)
- Topology-Aware Conformal Prediction for Stream Networks (NeurIPS 2025 · similarity 0.85)
- Dynamic Correlation Clustering in Sublinear Update Time (ICML 2024 · similarity 0.84)
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.
