Research map / Learning theory and algorithms / Kernels and regression theory
Solving Interpretable Kernel Dimensionality Reduction
Chieh Wu, Jared Miller, Yale Chang, Mario Sznaier, Jennifer Dy
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 |
|---|---|---|---|
| Replace the manifold optimizer with ISM | IKDR supervised and unsupervised optimization · Face | runtime: 1.92d -> 0.99s | Improved |
| Replace manifold optimization with ISM | IKDR supervised dimension reduction · Wine, Cancer, Face, MNIST | accuracy: 93.2\% $\pm$ 5.5\% -> 95.0\% $\pm$ 5\%; 97.3\%$\pm$ 0.3\% -> 97.3\%$\pm$ 0.3\% | Mixed |
| Replace Gaussian kernel with polynomial kernel (ablation) | ISM · Supervised dimension reduction | accuracy: 95.0\% $\pm$ 5\% -> \textbf{97.2\%} $\pm$ \textbf{3.7\%} | Mixed |
| Replace Gaussian kernel with multiquadratic kernel (ablation) | ISM · Unsupervised dimension reduction | NMI: 0.86 -> 0.88; 0.94 -> 0.92 | Mixed |
| Combine Gaussian and polynomial kernels using center alignment (ablation) | ISM · Supervised dimension reduction | accuracy: 97.2\% $\pm$ 3.7\% -> \textbf{98.3\% $\pm$ 2.6\%} | Improved |
| Use a squared kernel instead of a linear kernel (ablation) | ISM · MNIST supervised | accuracy: 99.1\% $\pm$ 0.1\% -> \textbf{99.3\% $\pm$ 0.2\%} | Improved |
| Use a multiquadratic kernel instead of a linear kernel (ablation) | ISM · Cancer unsupervised | NMI: 0.80 -> 0.84 | Improved |
| Use a squared kernel instead of a linear kernel (ablation) | ISM · Face unsupervised | NMI: 0.93 -> \textbf{0.95} | Improved |
Closest papers by meaning
- Learning in Feature Spaces via Coupled Covariances: Asymmetric Kernel SVD and Nyström method (ICML 2024 · similarity 0.85)
- Deep Principal Support Vector Machines for Nonlinear Sufficient Dimension Reduction (ICML 2025 · similarity 0.84)
- Random features for Grassmannian kernel approximation with bounded rank-one projections (NeurIPS 2026 · similarity 0.83)
- Statistical and Computational Guarantees of Kernel Max-Sliced Wasserstein Distances (ICML 2025 · similarity 0.83)
- Supervised Kernel Thinning (NeurIPS 2024 · similarity 0.83)
- Incremental Nyström-based Multiple Kernel Clustering (AAAI 2025 · 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.
