Research map / Learning theory and algorithms / Optimal transport
Sliced Wasserstein Regression
Han Chen, Yidong Zhou, Hans-Georg Müller
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 |
|---|---|---|---|
| use SWW (slice-wise) instead of SAW (slice-averaged) | SAW · global regression, Gaussian p=2 n=200 | ISE: 0.073 -> 0.037 | Improved |
| use SWW instead of SAW | SAW · local regression, Gaussian p=2 n=50 | ISE: 0.150 vs 0.126 | Hurt |
| use SWW instead of SAW | SAW · global regression, non-Gaussian p=5 n=200 | ISE: 0.071 -> 0.019 | Improved |
| use sliced regression (SWW) instead of FM (Sinkhorn) | FM · global Gaussian p=2 n=200 | ISE: 0.416 -> 0.037 | Improved |
| use sliced regression (SAW) instead of FM | FM · local non-Gaussian p=2 n=200 | ISE: 0.422 -> 0.073 | Improved |
| increase sample size n from 50 to 200 (ablation) | SWW · global Gaussian p=2 | ISE: 0.080 -> 0.037 | Improved |
| increase sample size n from 50 to 200 (ablation) | FM · global Gaussian p=2 | ISE: 0.460 -> 0.416 | Improved |
| use SWW instead of FM for runtime | FM · global p=2 n=200 | runtime (s): 285.62 -> 1.42 | Improved |
| use SWW instead of SAW for runtime | SAW · global p=5 n=200 | runtime (s): 130.45 -> 4.83 | Improved |
| make responses non-Gaussian via random transport map | SWW · global p=5 n=200 | ISE: 0.027 -> 0.019 | No clear effect |
Closest papers by meaning
- Functional data analysis for multivariate distributions through Wasserstein slicing (NeurIPS 2025 · similarity 0.88)
- Sliced Wasserstein Estimation with Control Variates (ICLR 2024 · similarity 0.86)
- Sliced Wasserstein with Random-Path Projecting Directions (ICML 2024 · similarity 0.84)
- Bayesian Additive Distribution Regression (NeurIPS 2026 · similarity 0.83)
- Risk and cross validation in ridge regression with correlated samples (ICML 2025 · similarity 0.83)
- Statistical and Computational Guarantees of Kernel Max-Sliced Wasserstein Distances (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.
