Research map / Probabilistic and scientific modeling / Variational and generative sampling
Energy-Based Sliced Wasserstein Distance
Khai Nguyen, Nhat Ho
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 uniform slicing (SW) with IS-EBSW-e energy-based slicing | sliced Wasserstein distance · point-cloud gradient flow, step 500 | Wasserstein-2 (x10^4): 9.41 -> 3.3 | Improved |
| replace Max-SW with IS-EBSW-e | sliced Wasserstein distance · point-cloud gradient flow, step 500 | Wasserstein-2 (x10^4): 6.83 -> 3.3 | Improved |
| replace v-DSW with IS-EBSW-e | sliced Wasserstein distance · point-cloud gradient flow, step 500 | Wasserstein-2 (x10^4): 5.68 -> 3.3 | Improved |
| replace Max-SW with IS-EBSW-e | sliced Wasserstein distance · point-cloud gradient flow | computation time (s): 28.38 -> 17.63 | Improved |
| replace v-DSW with IS-EBSW-e | sliced Wasserstein distance · point-cloud gradient flow | computation time (s): 21.2 -> 17.63 | Improved |
| replace SW with IS-EBSW-e | sliced Wasserstein distance · point-cloud gradient flow | computation time (s): 17.06 -> 17.63 | No clear effect |
| use exponential energy function instead of identity energy function (ablation) | EBSW · point-cloud gradient flow | Wasserstein-2 | Improved |
| use SIR, IMH or RMH sampling instead of importance sampling (ablation) | EBSW-e · point-cloud gradient flow | Wasserstein-2 and time | No clear effect |
| replace SW with IS-EBSW-e | point-cloud autoencoder · ModelNet40, epoch 20 | SW2: 2.97 -> 2.68 | Improved |
| replace SW with IS-EBSW-e | point-cloud autoencoder · ModelNet40, epoch 20 | W2: 12.67 -> 11.90 | Improved |
| replace Max-SW with IS-EBSW-e | point-cloud autoencoder · ModelNet40, epoch 200 | W2: 9.84 -> 9.69 | Improved |
| replace v-DSW with IS-EBSW-e | point-cloud autoencoder · ModelNet40, epoch 200 | SW2: 2.07 -> 2.04 | Improved |
Closest papers by meaning
- Sliced Wasserstein Estimation with Control Variates (ICLR 2024 · similarity 0.91)
- Sliced Wasserstein with Random-Path Projecting Directions (ICML 2024 · similarity 0.89)
- Quasi-Monte Carlo for 3D Sliced Wasserstein (ICLR 2024 · similarity 0.89)
- Tree-Sliced Wasserstein Distance with Nonlinear Projection (ICML 2025 · similarity 0.89)
- Towards Understanding Gradient Dynamics of the Sliced-Wasserstein Distance via Critical Point Analysis (ICML 2025 · similarity 0.89)
- Sliced-Wasserstein Estimation with Spherical Harmonics as Control Variates (ICML 2024 · similarity 0.88)
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.
