Research map / Learning theory and algorithms / Kernels and regression theory
Kernel Functional Optimisation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 KFO kernel instead of RBF | GP regression · UCI Boston | test negative log-likelihood: 139.61 ± 11.5 -> −24.7 ± 4.2 | Helped |
| use KFO kernel instead of ARD Matérn | GP regression · UCI Airfoil | test negative log-likelihood: 305.88 ± 7.46 -> −204.1±4.5 | Helped |
| use KFO kernel instead of FKL-Shared | GP regression · UCI Slump | test negative log-likelihood: 4.38 ± 1.33 -> −3.01 ± 1.7 | Helped |
| use KFO kernel instead of RBF/ARD Matérn | GP regression · UCI Fertility | test negative log-likelihood: −4.40 ± 2.58 -> 5.15 ± 2.95 | Hurt |
| use KFO kernel instead of SVM-RBF | SVM · UCI Sonar | test classification error (%): 15.3 ± 4.1 -> 6.6 ± 1.2 | Helped |
| use KFO kernel instead of SVM-MKL | SVM · UCI Ionosphere | test classification error (%): 9.31 ± 0.4 -> 3.06 ± 1.6 | Helped |
| use KFO kernel instead of SVM-RBF | SVM · UCI Credit | test classification error (%): 15.3 ± 2.0 -> 28.19 ± 3.57 | Hurt |
| use KFO kernel instead of SVM-RBF | SVM · UCI Glass | test classification error (%): 7.2 ± 2.7 -> 9.3 ± 0.19 | Hurt |
| use KFO kernel instead of SVM-RBF | SVM · UCI Seeds | test classification error (%): 8.74 ± 0.4 -> 1.2 ± 0.2 | Helped |
| KFO versus best classifiers in literature | SVM · 14 UCI datasets | datasets where best: 9/14 | Helped |
| use KFO kernel instead of SE kernel | GP · Triangular wave | max log-likelihood: −173 ± 1.2 -> 89.8 ± 1.6 | Helped |
| use KFO kernel instead of MKL | GP · Triangular wave | max log-likelihood: 82.1 ± 0.4 -> 89.8 ± 1.6 | Helped |
Closest papers by meaning
- Automated Kernel Discovery Towards Understanding High-dimensional Bayesian Optimization (NeurIPS 2026 · similarity 0.89)
- Kernel-based guarantees for nonlinear parametric models in Bayesian optimization (NeurIPS 2026 · similarity 0.86)
- The optimality of kernel classifiers in Sobolev space (ICLR 2024 · similarity 0.86)
- Scalable Kernel Inverse Optimization (NeurIPS 2024 · similarity 0.85)
- Adaptive Kernel Design for Bayesian Optimization Is a Piece of CAKE with LLMs (NeurIPS 2025 · similarity 0.85)
- On the Target-kernel Alignment: a Unified Analysis with Kernel Complexity (NeurIPS 2024 · 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.