Research map / Learning theory and algorithms / Kernels and regression theory
Koopman Kernel Regression
Petar Bevanda, Max Beier, Armin Lederer, Stefan Georg Sosnowski, Eyke Hüllermeier, Sandra Hirche
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 |
|---|---|---|---|
| Increase the number of eigenvalues from 16 to 200 with uniform spectral sampling (ablation) | KKR · Van der Pol, ℓ=10^1 | average risk [× 10^-2]: 13.7 -> 5.38 | Improved |
| Increase the number of eigenvalues from 16 to 200 with uniform spectral sampling (ablation) | KKR · Van der Pol, ℓ=10^0 | average risk [× 10^-2]: 6.46 -> 0.78 | Improved |
| Increase the number of eigenvalues from 16 to 200 with uniform spectral sampling (ablation) | KKR · Van der Pol, ℓ=10^-1 | average risk [× 10^-2]: 7.12 -> 1.74 | Improved |
| Use physics-informed rather than uniform spectral sampling at D=16 (ablation) | KKR · Van der Pol, ℓ=10^1 | average risk [× 10^-2]: 13.7 -> 5.60 | Improved |
| Use physics-informed rather than uniform spectral sampling at D=200 (ablation) | KKR · Van der Pol, ℓ=10^1 | average risk [× 10^-2]: 5.38 -> 5.58 | Mixed |
| Use physics-informed rather than uniform spectral sampling at D=16 (ablation) | KKR · Van der Pol, ℓ=10^0 | average risk [× 10^-2]: 6.46 -> 0.97 | Improved |
| Use physics-informed rather than uniform spectral sampling at D=200 (ablation) | KKR · Van der Pol, ℓ=10^0 | average risk [× 10^-2]: 0.78 -> 0.92 | Hurt |
| Use physics-informed rather than uniform spectral sampling at D=16 (ablation) | KKR · Van der Pol, ℓ=10^-1 | average risk [× 10^-2]: 7.12 -> 1.83 | Improved |
| Use physics-informed rather than uniform spectral sampling at D=200 (ablation) | KKR · Van der Pol, ℓ=10^-1 | average risk [× 10^-2]: 1.74 -> 1.80 | Mixed |
| Use trajectory-level forecasting-risk minimization rather than one-step operator regression | KKR versus PCR/EDMD · Bi-stable system | forecasting accuracy | Improved |
| Use KKR rather than PCR | Flow past a cylinder · Flow past a cylinder, 99-step forecast | forecast risk | Improved |
| Use KKR rather than Sig-PDE | Flow past a cylinder · Flow past a cylinder | forecast stability | Improved |
Closest papers by meaning
- SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting (ICML 2025 · similarity 0.88)
- Efficient Parametric SVD of Koopman Operator for Stochastic Dynamical Systems (NeurIPS 2025 · similarity 0.87)
- Deep-Koopman-KANDy: Dictionary Discovery for Deep-Koopman Operators with Kolmogorov-Arnold Networks for Dynamics (NeurIPS 2026 · similarity 0.87)
- Generalizing across Temporal Domains with Koopman Operators (AAAI 2024 · similarity 0.86)
- ResKoopNet: Learning Koopman Representations for Complex Dynamics with Spectral Residuals (ICML 2025 · similarity 0.86)
- Efficient Dynamics Modeling in Interactive Environments with Koopman Theory (ICLR 2024 · similarity 0.86)
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.
