Research map / Learning theory and algorithms / Kernels and regression theory
On lp-Support Vector Machines and Multidimensional Kernels
Victor Blanco, Justo Puerto, Antonio M. Rodriguez-Chia
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 l4/3-SVM with monomial transformation (eta=1) instead of RBF l2-SVM | RBF l2-SVM · housing | test accuracy (%): 85.35 -> 85.36 | No clear effect |
| use l3/2-SVM with Gaussian transformation (eta=3) instead of RBF l2-SVM | RBF l2-SVM · german credit | test accuracy (%): 77.30 -> 79.00 | Improved |
| use l4/3-SVM (eta=1) instead of RBF l2-SVM | RBF l2-SVM · colon | test accuracy (%): 80.24 -> 82.14 | Improved |
| use l4/3-SVM with monomial transformation eta=4 instead of RBF l2-SVM | RBF l2-SVM · page blocks | test accuracy (%): 92.31 -> 96.11 | Improved |
| use lp-SVM instead of RBF l2-SVM | RBF l2-SVM · cleveland | test accuracy (%): 85.16 -> 83.48 | Hurt |
| raise monomial degree eta from 1 to 3 (ablation) | l4/3-SVM, monomial transformation · cleveland | test accuracy (%): 82.84 -> 74.93 | Hurt |
| raise monomial degree eta from 1 to 4 (ablation) | l4/3-SVM, monomial transformation · page blocks | test accuracy (%): 93.49 -> 96.11 | Improved |
| use Gaussian instead of monomial transformation at eta=4 (ablation) | l4/3-SVM · cleveland | test accuracy (%): 76.56 -> 83.47 | Improved |
| use Gaussian instead of monomial transformation at eta=3 (ablation) | l4/3-SVM · german credit | test accuracy (%): 71.90 -> 76.75 | Improved |
| use l4/3 norm instead of l2 norm (ablation) | SVM, Gaussian transformation (eta=2) · cleveland | test accuracy (%): 80.58 -> 84.19 | Improved |
| use l4/3 norm instead of l2 norm (ablation) | SVM, Gaussian transformation (eta=4) · cleveland | nonzero coefficients (%): 41.97% -> 3.53% | Improved |
| use l2 norm instead of lp norm (solve time) (ablation) | SVM, Gaussian transformation (eta=4) · cleveland | CPU time (s): 19.57 -> 0.19 | Improved |
Closest papers by meaning
- Solving L1-regularized SVMs and Related Linear Programs: Revisiting the Effectiveness of Column and Constraint Generation (JMLR 2022 · similarity 0.90)
- A Unified View on Multi-class Support Vector Classification (JMLR 2016 · similarity 0.88)
- An Error Bound for L1-norm Support Vector Machine Coefficients in Ultra-high Dimension (JMLR 2016 · similarity 0.87)
- Faster Algorithms for Structured Linear and Kernel Support Vector Machines (ICLR 2025 · similarity 0.87)
- Support vector machines and linear regression coincide with very high-dimensional features (NeurIPS 2021 · similarity 0.87)
- Deep Principal Support Vector Machines for Nonlinear Sufficient Dimension Reduction (ICML 2025 · similarity 0.85)
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.
