Research map / Learning theory and algorithms / Clustering
A Sampling-Based Approach for Efficient Clustering in Large Datasets
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 |
|---|---|---|---|
| Use D-GMM instead of vc-GMM | GMM clustering · KDD | distance evaluation speedup: \times533.1\pm36.5 -> \times622.1\pm28.0 | Helped |
| Use D-GMM instead of vc-GMM | GMM clustering · KDD | relative quantisation error: 12.0\pm0.4\% -> 12.0\pm1.0\% | No clear effect |
| Use D-GMM instead of vc-GMM | GMM clustering · CIFAR-10 | distance evaluation speedup: \times674.7\pm45.8 -> \times731.5\pm41.9 | Helped |
| Use D-GMM instead of vc-GMM | GMM clustering · CIFAR-10 | relative quantisation error: 7.0\pm0.0\% -> 8.0\pm0.0\% | Hurt |
| Use D-GMM instead of vc-GMM | GMM clustering · SONG | distance evaluation speedup: \times698.2\pm0.7 -> \times862.1\pm18.3 | Helped |
| Use D-GMM instead of vc-GMM | GMM clustering · SONG | relative quantisation error: 8.0\pm0.1\% -> 8.0\pm0.2\% | No clear effect |
| Use D-GMM instead of vc-GMM | GMM clustering · SUSY | distance evaluation speedup: \times663.1\pm17.1 -> \times605.7\pm11.1 | Hurt |
| Use D-GMM instead of vc-GMM | GMM clustering · SUSY | relative quantisation error: 6.0\pm0.1\% -> 5.0\pm0.1\% | Helped |
| Increase the coreset size (ablation) | D-GMM versus vc-GMM on ImageNet · ImageNet 64x64 | reduction in distance evaluations versus vc-GMM: 22\%, 31\%, 40\%, 44\%, and 49\% | Helped |
| Reduce search parameter R while fixing H=5 (ablation) | D-GMM · CIFAR-10 hyperparameter search | number of operations | Helped |
| Set R below 40 while fixing H=5 (ablation) | D-GMM · CIFAR-10 hyperparameter search | runtime speedup | Hurt |
| Reduce truncated-space size H while fixing R=10 (ablation) | D-GMM · CIFAR-10 hyperparameter search | runtime | Helped |
Closest papers by meaning
- Efficient Parameter-Free Clustering Using First Neighbor Relations (CVPR 2019 · similarity 0.92)
- Simple, Scalable and Effective Clustering via One-Dimensional Projections (NeurIPS 2023 · similarity 0.91)
- Efficient Clustering Based On A Unified View Of $K$-means And Ratio-cut (NeurIPS 2020 · similarity 0.91)
- Clustering Small Samples With Quality Guarantees: Adaptivity With One2all PPS (AAAI 2018 · similarity 0.91)
- Ultrametric Cluster Hierarchies: I Want ‘em All! (NeurIPS 2025 · similarity 0.90)
- Near-Linear Time Approximation Algorithms for k-means with Outliers (ICML 2024 · 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.