Research map / Learning theory and algorithms / Clustering
Supervised Hierarchical Clustering with Exponential Linkage
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 |
|---|---|---|---|
| Replace all-pairs training with MST-based training | Single-linkage HAC · Faces, NP Coref, Rexa, AMINER | dendrogram purity | Helped |
| Match the training procedure to the linkage function | Supervised HAC · Four datasets | dendrogram purity: up to 8 points | Helped |
| Jointly learn the exponential-linkage parameter and pairwise dissimilarity | Exponential Linkage HAC · AMINER | dendrogram purity | Helped |
| Jointly learn the exponential-linkage parameter and pairwise dissimilarity | Exponential Linkage HAC · NP Coref | dendrogram purity | Helped |
| Jointly learn the exponential-linkage parameter and pairwise dissimilarity rather than use all-pairs training | Exponential Linkage HAC · Rexa | pairwise F1: more than 20 points higher F1 | Helped |
| Use all-pairs training instead of the best-performing training procedure | Single-linkage and Exponential Linkage HAC · Rexa, NP Coref, Faces | dendrogram purity: 4-8% | Hurt |
| Use all-pairs training instead of the best-performing training procedure | Average-linkage HAC · Four datasets | dendrogram purity | Hurt |
| Use complete linkage rather than another tested linkage | Supervised HAC · Four datasets | dendrogram purity | Hurt |
| Train with expAlpha rather than all-pairs | Exponential Linkage HAC · Rexa | pairwise F1: more than 20 points higher F1 | Helped |
| Use the best-performing method rather than expPlus | Supervised HAC · AMINER | pairwise F1: ~20 points F1 | Helped |
| Use exponential linkage with alpha = -1 instead of single, average, or complete linkage | HAC with Euclidean distance · Synthetic three-cluster dataset | cluster recovery | Helped |
| Choose an exponential-linkage member after training the dissimilarity function rather than fixing a standard linkage beforehand | Supervised HAC · Rexa, Faces | dendrogram purity | Mixed |
Closest papers by meaning
- Average Sensitivity of Hierarchical $k$-Median Clustering (ICML 2025 · similarity 0.88)
- Classification Fields: Arbitrarily Fine Recursive Hierarchical Clustering From Few Examples (NeurIPS 2026 · similarity 0.87)
- New Bounds on the Cohesion of Complete-link and Other Linkage Methods for Agglomerative Clustering (ICML 2024 · similarity 0.87)
- From Logits to Hierarchies: Hierarchical Clustering made Simple (ICML 2025 · similarity 0.86)
- On the cohesion and separability of average-link for hierarchical agglomerative clustering (NeurIPS 2024 · similarity 0.86)
- Ultrametric Cluster Hierarchies: I Want ‘em All! (NeurIPS 2025 · similarity 0.86)
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.