Research map / Learning theory and algorithms / Clustering
Scalable Algorithm for Higher-Order Co-Clustering via Random Sampling
Daisuke Hatano, Takuro Fukunaga, Takanori Maehara, Ken-ichi Kawarabayashi
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 distorted sampling heuristic (ablation) | Prop · Synthetic e,3,100,3 | NMI: 0.06 -> 0.77 | Improved |
| add distorted sampling heuristic (ablation) | Prop · Synthetic e,4,50,4 | NMI: 0.63 -> 0.95 | Improved |
| add distorted sampling heuristic (ablation) | Prop · Synthetic e,5,50,5 | NMI: 0.03 -> 0.78 | Improved |
| add distorted sampling heuristic (ablation) | Prop · Synthetic u,3,100,4 | NMI: 0.60 -> 0.07 | Hurt |
| add balancing merge heuristic (ablation) | Prop · Synthetic e,3,100,3 | NMI: 0.06 -> 0.09 | Improved |
| add balancing merge heuristic (ablation) | Prop · Synthetic u,3,100,5 | NMI: 0.04 -> 0.17 | Improved |
| add both heuristics (ablation) | Prop · Synthetic u,3,100,4 | NMI: 0.60 -> 0.93 | Improved |
| add both heuristics rather than distorted sampling alone (ablation) | Prop+D · Synthetic u,3,100,4 | NMI: 0.07 -> 0.93 | Improved |
| Prop+D+B vs PARAFAC on order-5 tensors | Prop+D+B · Synthetic e,5,50,4 | runtime (s): 282 -> 76 | Improved |
| Prop+D+B vs PARAFAC on higher-order tensors | Prop+D+B · Synthetic e,5,50,5 | NMI: 0.79 -> 0.79 | No clear effect |
| Prop+D+B vs PARAFAC on Amazon 778^3 tensor | Prop+D+B · Amazon review tensors | runtime (s): 1039 -> 11.5 | Improved |
| Prop+D+B vs PARAFAC on order-3 small tensors | Prop+D+B · Synthetic e,3,100,3 | runtime (s): 1.7 -> 5.1 | Hurt |
Closest papers by meaning
- General Tensor Spectral Co-clustering for Higher-Order Data (NeurIPS 2016 · similarity 0.91)
- Uniform Hypergraph Partitioning: Provable Tensor Methods and Sampling Techniques (JMLR 2017 · similarity 0.87)
- Memory-Efficient Approximation Algorithms for Max-k-Cut and Correlation Clustering (NeurIPS 2021 · similarity 0.87)
- Fast Combinatorial Algorithms for Min Max Correlation Clustering (ICML 2023 · similarity 0.87)
- Edge-Colored Clustering in Hypergraphs: Beyond Minimizing Unsatisfied Edges (ICML 2025 · similarity 0.85)
- A Near-Linear Time Approximation Algorithm for Beyond-Worst-Case Graph Clustering (ICML 2024 · 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.
