Research map / Learning theory and algorithms / Clustering
Multidimensional Scaling on Multiple Input Distance Matrices
Song Bai, Xiang Bai, Longin Jan Latecki, Qi Tian
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 |
|---|---|---|---|
| learn view weights adaptively instead of using an equal-weight linear combination | MVMDS · MSRC-v1 | NN retrieval: 0.796 -> 0.806 | Improved |
| learn view weights adaptively instead of using an equal-weight linear combination | MVMDS · Caltech101-7 | FT retrieval: 0.460 -> 0.554 | Improved |
| learn view weights adaptively instead of using an equal-weight linear combination | MVMDS · Caltech101-20 | NN retrieval: 0.304 -> 0.429 | Improved |
| learn view weights adaptively instead of using an equal-weight linear combination | MVMDS · Caltech101-7 | ACC (%) clustering: 56.7$\pm$2.68 -> 71.5$\pm$1.64 | Improved |
| learn view weights adaptively instead of using an equal-weight linear combination | MVMDS · Caltech101-7 | NMI (%) clustering: 44.8$\pm$2.98 -> 63.0$\pm$1.97 | Improved |
| learn view weights adaptively instead of using an equal-weight linear combination | MVMDS · Caltech101-7 | Purity (%) clustering: 63.2$\pm$2.44 -> 76.1$\pm$1.32 | Improved |
| use MVMDS with $\gamma$=5 rather than LC_MDS | six-city synthetic example · Synthetic six-city example | Stress: 6.15 -> 1.35 | Improved |
| set the weight controller to $\gamma$=1.5 instead of $\gamma$=5 (ablation) | MVMDS · Synthetic six-city example | Stress: 1.35 -> 1.61 | Hurt |
| set the weight controller to $\gamma$=10 instead of $\gamma$=5 (ablation) | MVMDS · Synthetic six-city example | Stress: 1.35 -> 1.36 | No clear effect |
| use an equal-weight linear combination of views | MDS · Caltech101-7 | NN retrieval: 0.788 -> 0.711 | Hurt |
| use an equal-weight linear combination of views | MDS · Caltech101-20 | NN retrieval: 0.417 -> 0.304 | Hurt |
Closest papers by meaning
- Multidimensional Scaling: Approximation and Complexity (ICML 2021 · similarity 0.90)
- Neuc-MDS: Non-Euclidean Multidimensional Scaling Through Bilinear Forms (NeurIPS 2024 · similarity 0.88)
- Finsler Multi-Dimensional Scaling: Manifold Learning for Asymmetric Dimensionality Reduction and Embedding (CVPR 2025 · similarity 0.83)
- Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering (CVPR 2025 · similarity 0.83)
- Generalized Dimension Reduction Using Semi-Relaxed Gromov-Wasserstein Distance (AAAI 2025 · similarity 0.83)
- From Dictionary to Tensor: A Scalable Multi-View Subspace Clustering Framework with Triple Information Enhancement (NeurIPS 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.
