Research map / Learning theory and algorithms / Prediction and decision losses
Learning Distance for Sequences by Learning a Ground Metric
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 |
|---|---|---|---|
| learn ground metric with RVSML instead of no metric learning (DTW) | 1-NN with DTW · ChaLearn | accuracy (%): 61.12 -> 87.38 | Helped |
| learn ground metric with RVSML (OPW) | 1-NN with OPW · ChaLearn | accuracy (%): 59.38 -> 83.82 | Helped |
| learn ground metric with RVSML (OPW) | 1-NN with OPW · SAD | MAP (%): 59.77 -> 65.63 | Helped |
| learn ground metric with RVSML (DTW) | 1-NN with DTW · MSR Action3D | accuracy (%): 81.32 -> 82.78 | Helped |
| learn ground metric with RVSML (OPW) | 1-NN with OPW · MSR Action3D | accuracy (%): 84.25 -> 76.56 | Hurt |
| learn ground metric with RVSML (OPW) | 1-NN with OPW · MSR Activity3D | accuracy (%): 58.13 -> 57.50 | Hurt |
| learn ground metric with RVSML (DTW) | 1-NN with DTW · SAD | accuracy (%): 96.36 -> 96.23 | No clear effect |
| RVSML instead of LDMLT with DTW | 1-NN with DTW · MSR Action3D | accuracy (%): 84.98 -> 82.78 | Hurt |
| RVSML instead of generalized RVML (single-element virtual sequences) | 1-NN with DTW · ChaLearn | MAP (%): 31.21 -> 33.83 | Helped |
| conventional vector metric learning (ITML, LMNN, SCML) applied to sequences | 1-NN with DTW · ChaLearn | accuracy (%): 61.12 -> 52.17 | Mixed |
| RVSML (DTW) vs LDMLT in training time | metric learning · ChaLearn | training time: 213921.7 -> 2477.76 | Helped |
| RVSML (OPW) vs SWMD in training time | metric learning · SAD | training time: 7756.72 -> 208.67 | Helped |
Closest papers by meaning
- Order-Preserving Wasserstein Distance for Sequence Matching (CVPR 2017 · similarity 0.85)
- Curvilinear Distance Metric Learning (NeurIPS 2019 · similarity 0.85)
- Neural Distance Embeddings for Biological Sequences (NeurIPS 2021 · similarity 0.84)
- Learning Low-Dimensional Metrics (NeurIPS 2017 · similarity 0.84)
- Learning Distances from Data with Normalizing Flows and Score Matching (ICML 2025 · similarity 0.84)
- Volume-Aware Distance for Robust Similarity Learning (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.