Research map / Learning theory and algorithms / Kernels and regression theory
CosNet: A Generalized Spectral Kernel Network
Yanfang Xue, Pengfei Fang, Jinyue Tian, Shipeng Zhu, hui xue
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 cosine/sine initialization of complex weight matrix (ablation) | CosNet · Earthquakes | accuracy (%): 69.93 -> 71.76 | Improved |
| add cosine/sine initialization of complex weight matrix (ablation) | CosNet · DistalPhalanxTW | accuracy (%): 63.02 -> 63.60 | Improved |
| add cosine/sine initialization of complex weight matrix (ablation) | CosNet · Strawberry | accuracy (%): 96.80 -> 97.22 | Improved |
| add cosine/sine initialization of complex weight matrix (ablation) | CosNet · power regression | MSE: 0.8795 -> 0.8229 | Improved |
| add cosine/sine initialization of complex weight matrix (ablation) | CosNet · concreat regression | MSE: 1.3731 -> 1.3606 | Improved |
| add cosine/sine initialization of complex weight matrix (ablation) | CosNet · yacht regression | MSE: 3.7932 -> 3.6270 | Improved |
| use CosNet instead of DSKN (real-valued spectral kernel network) | DSKN · Wine | accuracy (%): 76.48 -> 85.46 | Improved |
| use CosNet instead of DSKN | DSKN · FordB | accuracy (%): 69.81 -> 71.73 | Improved |
| widen SRFF (more neurons per layer) (ablation) | SRFF · ECG200 | accuracy (%): 73.40 -> 83.90 | Improved |
| widen DSKN (more neurons per layer) (ablation) | DSKN · ECG200 | accuracy (%): 77.80 -> 80.65 | Improved |
| deepen SRFF (more layers) (ablation) | SRFF · ECG200 | accuracy (%): 73.40 -> 65.85 | Hurt |
| widen SRFF (ablation) | SRFF · Ham | accuracy (%): 68.43 -> 70.52 | Improved |
Closest papers by meaning
- Nonstationary Sparse Spectral Permanental Process (NeurIPS 2024 · similarity 0.86)
- Spectral Transformer Neural Processes (NeurIPS 2026 · similarity 0.84)
- Generalization error of spectral algorithms (ICLR 2024 · similarity 0.83)
- Learning Robust Spectral Dynamics for Temporal Domain Generalization (NeurIPS 2025 · similarity 0.83)
- Position: Spectral GNNs Rely Less on Graph Fourier Basis than Conceived (ICML 2025 · similarity 0.83)
- A General Graph Spectral Wavelet Convolution via Chebyshev Order Decomposition (ICML 2025 · 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.
