Research map / Learning theory and algorithms / Optimal transport
Order-Preserving Wasserstein Distance for Sequence Matching
Check 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 |
|---|---|---|---|
| Add OPW temporal regularization to Sinkhorn optimal transport | Sequence distance · Similar online Chinese characters | accuracy (NM): 34.97 (1.84) -> 57.75 (2.62) | Helped |
| Add OPW temporal regularization to Sinkhorn optimal transport | Sequence distance · Spoken Arabic Digits | accuracy (NM): 77.00 -> 87.27 | Helped |
| Use OPW instead of DTW | MSR Daily Activity3D with NM classifier · MSR Daily Activity3D | accuracy: 40.63 -> 38.75 | Hurt |
| Use OPW instead of DTW | MSR Daily Activity3D with 1-NN classifier · MSR Daily Activity3D | accuracy: 58.75 -> 58.13 | Hurt |
| Increase prior bandwidth σ beyond 5 (ablation) | OPW on MSR Action3D · MSR Action3D | accuracy | Hurt |
| Increase the KL-prior regularization weight λ2 (ablation) | OPW on MSR Action3D · MSR Action3D | performance | Hurt |
| Increase inverse difference moment weight λ1 while λ1 > 1 (ablation) | OPW on MSR Action3D · MSR Action3D | performance | No clear effect |
| Use a large λ1 with λ2 ≤ 0.01 (ablation) | OPW on MSR Action3D · MSR Action3D | performance | Hurt |
Closest papers by meaning
- A New Robust Partial p-Wasserstein-Based Metric for Comparing Distributions (ICML 2024 · similarity 0.86)
- Learning Distance for Sequences by Learning a Ground Metric (ICML 2019 · similarity 0.85)
- Drop-DTW: Aligning Common Signal Between Sequences While Dropping Outliers (NeurIPS 2021 · similarity 0.85)
- Scalable Approximation Algorithms for $p$-Wasserstein Distance and Its Variants (ICML 2025 · similarity 0.85)
- Wasserstein Wormhole: Scalable Optimal Transport Distance with Transformer (ICML 2024 · similarity 0.83)
- Discovering Sequential Patterns with Predictable Inter-event Delays (AAAI 2024 · 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.