Research map / Learning theory and algorithms / Kernels and regression theory
On the Similarity between the Laplace and Neural Tangent Kernels
Amnon Geifman, Abhay Yadav, Yoni Kasten, Meirav Galun, David Jacobs, Basri Ronen
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 NTK with homogeneous Laplace kernel | kernel regression on UCI datasets · 90 UCI datasets | Friedman rank/accuracy metrics | No clear effect |
| replace NTK with homogeneous gamma-exponential kernel | kernel regression on UCI datasets · 90 UCI datasets | Friedman rank/accuracy metrics | Improved |
| use homogeneous versions of exponential kernels instead of standard (ablation) | exponential kernels · 90 UCI datasets | classification metrics | Improved |
| replace NTK with H-Laplace | FALKON kernel methods · MillionSongs | MSE: 79.9666 -> 79.7941 | Improved |
| replace NTK with H-Laplace | FALKON kernel methods · HIGGS | AUC: 82.089 -> 81.995 | Hurt |
| replace NTK with H-gamma-exponential | FALKON kernel methods · HIGGS | AUC: 82.089 -> 82.281 | Improved |
| replace CNTK with C-Exp Laplace | hierarchical convolutional kernel · CIFAR-10 | accuracy: 66.4 -> 65.2 (50k); 43.9 -> 44.2 (2k) | Mixed |
| replace CNTK with C-Exp gamma-exponential | hierarchical convolutional kernel · CIFAR-10 | accuracy: 66.4 -> 67.0 (50k); 43.9 -> 45.2 (2k) | Improved |
| use Gaussian kernel instead of Laplace for eigenvalue decay | kernels on hypersphere · spectral analysis | eigenvalue decay | Hurt |
Closest papers by meaning
- Deep Neural Tangent Kernel and Laplace Kernel Have the Same RKHS (ICLR 2021 · similarity 0.92)
- Spectral Analysis of the Neural Tangent Kernel for Deep Residual Networks (JMLR 2024 · similarity 0.90)
- Divergence of Neural Tangent Kernel in Classification Problems (ICLR 2025 · similarity 0.88)
- Neural Tangent Kernels Motivate Cross-Covariance Graphs in Neural Networks (ICML 2024 · similarity 0.87)
- On the Impacts of the Random Initialization in the Neural Tangent Kernel Theory (NeurIPS 2024 · similarity 0.87)
- Optimized Random Features for the Neural Tangent Kernel (Student Abstract) (AAAI 2025 · similarity 0.87)
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.
