Research map / Learning theory and algorithms / Kernels and regression theory
Memory and Computation-Efficient Kernel SVM via Binary Embedding and Ternary Model Coefficients
Zijian Lei, Liang Lan
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 ternary rather than binary model coefficients (ablation) | Our Method · usps | accuracy: 96.57 -> 98.01 | Improved |
| use ternary rather than binary model coefficients (ablation) | Our Method · covtype | accuracy: 78.12 -> 81.68 | Improved |
| use ternary rather than binary model coefficients (ablation) | Our Method · webspam | accuracy: 94.75 -> 96.34 | Improved |
| use ternary rather than binary model coefficients (ablation) | Our Method · mnist | accuracy: 92.66 -> 93.27 | Improved |
| use ternary rather than binary model coefficients (ablation) | Our Method · fashion-mnist | accuracy: 82.07 -> 83.29 | Improved |
| use ternary rather than binary model coefficients (ablation) | Our Method · usps | memory cost: 29KB -> 104KB | Hurt |
| use ternary rather than binary model coefficients (ablation) | Our Method · covtype | memory cost: 24KB -> 32KB | Hurt |
| use ternary rather than binary model coefficients (ablation) | Our Method · webspam | memory cost: 24KB -> 32KB | Hurt |
| use ternary rather than binary model coefficients (ablation) | Our Method · mnist | memory cost: 29KB -> 104KB | Hurt |
| use ternary rather than binary model coefficients (ablation) | Our Method · fashion-mnist | memory cost: 29KB -> 104KB | Hurt |
| initialize coefficients from a linear SVM solution instead of random initialization (ablation) | Our Method · usps | accuracy | Improved |
| initialize coefficients from a linear SVM solution instead of random initialization (ablation) | Our Method · usps | objective value | Improved |
Closest papers by meaning
- Approximation Vector Machines for Large-scale Online Learning (JMLR 2017 · similarity 0.85)
- Memory Efficient Kernel Approximation (JMLR 2017 · similarity 0.85)
- Joker: Joint Optimization Framework for Lightweight Kernel Machines (ICML 2025 · similarity 0.85)
- Fast Support Vector Classification for Large-Scale Problems (TPAMI 2021 · similarity 0.84)
- Faster Algorithms for Structured Linear and Kernel Support Vector Machines (ICLR 2025 · similarity 0.84)
- Non-parametric Representation Learning with Kernels (AAAI 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.
