Research map / Learning theory and algorithms / Clustering
Twin Learning for Similarity and Clustering: A Unified Kernel Approach
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 |
|---|---|---|---|
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · YALE | Accuracy (%): 55.85 -> 56.97 | Helped |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · JAFFE | Accuracy (%): 99.83 -> 100.00 | Helped |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · ORL | Accuracy (%): 62.35 -> 65.25 | Helped |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · AR | Accuracy (%): 56.79 -> 62.38 | Helped |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · BA | Accuracy (%): 47.72 -> 47.34 | Hurt |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · TR11 | Accuracy (%): 71.26 -> 73.43 | Helped |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · TR41 | Accuracy (%): 63.52 -> 67.31 | Mixed |
| Jointly learn similarity, cluster indicators, and a weighted combination of kernels instead of using single-kernel SCSK | SCMK · TR45 | Accuracy (%): 74.02 -> 74.35 | Helped |
| Use the average over 12 kernels rather than the best kernel | SCSK · YALE | Accuracy (%): 55.85 -> 45.35 | Hurt |
| Use the average over 12 kernels rather than the best kernel | SCSK · JAFFE | Accuracy (%): 99.83 -> 86.64 | Hurt |
| Use the average over 12 kernels rather than the best kernel | SCSK · ORL | Accuracy (%): 62.35 -> 50.50 | Hurt |
| Use the average over 12 kernels rather than the best kernel | SCSK · AR | Accuracy (%): 56.79 -> 41.35 | Hurt |
Closest papers by meaning
- Unified Spectral Clustering With Optimal Graph (AAAI 2018 · similarity 0.92)
- Hierarchical Multiple Kernel Clustering (AAAI 2021 · similarity 0.90)
- Scalable Multiple Kernel Clustering: Learning Clustering Structure from Expectation (ICML 2024 · similarity 0.89)
- COKE: Core Kernel for More Efficient Approximation of Kernel Weights in Multiple Kernel Clustering (ICML 2025 · similarity 0.88)
- Explaining Kernel Clustering via Decision Trees (ICLR 2024 · similarity 0.86)
- Towards an Explainable Comparison and Alignment of Feature Embeddings (ICML 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.