Research map / Graph learning / Graph algorithms and clustering
The Product Cut
Thomas Laurent, James von Brecht, Xavier Bresson, arthur szlam
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 |
|---|---|---|---|
| use Product Cut with AMG algorithm instead of Normalized Cut spectral (NCUT) | PCut · 20NEWS | cluster purity: 27 -> 61 | Improved |
| use PCut instead of NCUT | PCut · RCV1 | cluster purity: 38 -> 53 | Improved |
| use PCut instead of NCUT | PCut · MNIST | cluster purity: 77 -> 97 | Improved |
| use PCut instead of NCUT | PCut · OPTDIGITS | cluster purity: 91 -> 98 | Improved |
| use PCut instead of GRACLUS | PCut · 20NEWS | cluster purity: 42 -> 61 | Improved |
| use PCut instead of MTV | PCut · 20NEWS | cluster purity: 36 -> 61 | Improved |
| use PCut instead of NMFR | PCut · RCV1 | cluster purity: 43 -> 53 | Improved |
| use PCut instead of NMFR | PCut · 20NEWS | cluster purity: 61 -> 61 | No clear effect |
| use PCut instead of NMFR | PCut · MNIST | cluster purity: 97 -> 97 | No clear effect |
| use PCut instead of GRACLUS/MTV on digit data | PCut · MNIST | cluster purity: 97 -> 97 | No clear effect |
| use PCut instead of NMFR for speed | PCut (lambda1) · MNIST | time to 95% of limiting purity: 4.6mn -> 11s | Improved |
| use PCut instead of NMFR for speed | PCut (lambda1) · 20NEWS | time to 95% of limiting purity: 3.7mn -> 1.3mn | Improved |
Closest papers by meaning
- Joint Cuts and Matching of Partitions in One Graph (CVPR 2018 · similarity 0.88)
- Analysis and Optimization of Graph Decompositions by Lifted Multicuts (ICML 2017 · similarity 0.86)
- A Near-Linear Time Approximation Algorithm for Beyond-Worst-Case Graph Clustering (ICML 2024 · similarity 0.84)
- Continuous Partitioning for Graph-Based Semi-Supervised Learning (NeurIPS 2024 · similarity 0.83)
- Edge-Colored Clustering in Hypergraphs: Beyond Minimizing Unsatisfied Edges (ICML 2025 · similarity 0.83)
- Combinatorial Approximations for Cluster Deletion: Simpler, Faster, and Better (ICML 2024 · similarity 0.83)
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.
