Research map / Learning theory and algorithms / Optimal transport
Distribution Regression with Sliced Wasserstein Kernels
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 a Gaussian SW2 kernel instead of doubly Gaussian MMD | Distribution-regression KRR · Synthetic mode classification, all reported configurations | RMSE: 0.64 ~(0.09) -> 0.48 ~(0.03); 3.66 ~(0.88) -> 2.92 ~(0.32); 0.68 ~(0.02) -> 0.54 ~(0.05); 3.48 ~(0.95) -> 3.17 ~(0.32); | Helped |
| Use a Gaussian SW2 kernel instead of RBF on pixel vectors | KRR image classification · MNIST raw | accuracy: 0.9 ~(0.01) -> 0.93 ~(0.01) | Helped |
| Use a Gaussian SW2 kernel instead of RBF on pixel vectors | KRR image classification · MNIST, max rotation π/12 and π/6 | accuracy: 0.51 ~(0.03) -> 0.85 ~(0.02); 0.47 ~(0.03) -> 0.82 ~(0.01) | Helped |
| Use a Gaussian SW2 kernel instead of RBF on pixel vectors | KRR image classification · Fashion MNIST raw | accuracy: 0.82~( 0.01) -> 0.81 ~(0.01) | Hurt |
| Use a Gaussian SW2 kernel instead of RBF on pixel vectors | KRR image classification · Fashion MNIST, max rotation π/12 and π/6 | accuracy: 0.66 (0.01) -> 0.74 ~(0.01); 0.64~(0.01) -> 0.70~(0.01) | Helped |
| Use SW1 instead of SW2 | Gaussian Sliced-Wasserstein KRR · Synthetic mode classification | RMSE | Mixed |
Closest papers by meaning
- Kernel Quantile Embeddings and Associated Probability Metrics (ICML 2025 · similarity 0.89)
- On Statistical Learning Theory for Distributional Inputs (ICML 2024 · similarity 0.89)
- Statistical and Computational Guarantees of Kernel Max-Sliced Wasserstein Distances (ICML 2025 · similarity 0.89)
- Learning to Embed Distributions via Maximum Kernel Entropy (NeurIPS 2024 · similarity 0.88)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.87)
- Fourier Sliced-Wasserstein Embedding for Multisets and Measures (ICLR 2025 · 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.