Research map / Learning theory and algorithms / Kernels and regression theory
Nonlinear Pairwise Layer and Its Training for Kernel Learning
Fanghui Liu, Xiaolin Huang, Chen Gong, Jie Yang, Li Li
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 |
|---|---|---|---|
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · 2-D toy sandwich data | training accuracy: 86.87% -> 94.93% | Improved |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI diabetic | test accuracy: 73.00±1.74 -> 81.98±1.72 | Improved |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI haberman | test accuracy: 73.26±3.89 -> 85.81±4.28 | Improved |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI monks, spambase, EEG | test accuracy: about 1%~3% | Improved |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI breast-cancer | test accuracy: 96.53±0.84 -> 97.35±0.74 | Hurt |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI monks2 | test accuracy: 85.81±1.40 -> 83.33±1.68 | Hurt |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI heart | training accuracy: 88.96±3.07 -> 92.59±4.97 | Improved |
| add nonlinear pairwise layer (KNPL vs SVM-CV) | SVM · UCI SPECT | training accuracy: 87.00±3.78 -> 92.75±6.99 | Improved |
| KNPL vs multiple kernel methods (DMKL, BMKL, EasyMKL) | SVM · 15 UCI datasets | test accuracy rank | Improved |
| KNPL vs EasyMKL | SVM · UCI breast-cancer | test accuracy: 99.52±0.13 -> 97.35±0.74 | Hurt |
| ADMM for pairwise matrix G | SVM · heart | iterations to converge: about 3 | Improved |
| alternating optimization of alpha (Algorithm 2) | SVM · heart | iterations to converge: within 10 iterations | Improved |
Closest papers by meaning
- Joker: Joint Optimization Framework for Lightweight Kernel Machines (ICML 2025 · similarity 0.88)
- Trusting SVM for Piecewise Linear CNNs (ICLR 2017 · similarity 0.87)
- Limited Memory Online Gradient Descent for Kernelized Pairwise Learning with Dynamic Averaging (AAAI 2024 · similarity 0.85)
- On the Target-kernel Alignment: a Unified Analysis with Kernel Complexity (NeurIPS 2024 · similarity 0.84)
- Deep Principal Support Vector Machines for Nonlinear Sufficient Dimension Reduction (ICML 2025 · similarity 0.84)
- Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization (NeurIPS 2025 · similarity 0.84)
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.
