Research map / Label-efficient and robust learning / Recognition and long-tail
EMD Metric Learning
Zizhao Zhang, Yubo Zhang, Xibin Zhao, Yue Gao
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 |
|---|---|---|---|
| Learn EMD metric (EMD-M) instead of traditional EMD, 20% training | EMD · TWITTER | accuracy: +5.48% | Improved |
| Learn EMD metric (EMD-M) instead of traditional EMD, 30% training | EMD · TWITTER | accuracy: +6.83% | Improved |
| Learn EMD metric (EMD-M) instead of traditional EMD, 40% training | EMD · TWITTER | accuracy: +8.32% | Improved |
| Learn EMD metric (EMD-M) instead of traditional EMD, 50% training | EMD · TWITTER | accuracy: +8.5% | Improved |
| EMD-M versus PML with 50% training data | PML · TWITTER | accuracy: +5.29% | Improved |
| EMD-M versus CDL with 50% training data | CDL · TWITTER | accuracy: +3.98% | Improved |
| EMD-M versus CDL_PLS with 50% training data | CDL_PLS · TWITTER | accuracy: +2.74% | Improved |
| Increase number of imposters ki beyond about 10 (ablation) | EMD-M · NTU | accuracy | No clear effect |
| Set lambda away from 0.5 (NTU) or 0.2 (TWITTER), too small or too large (ablation) | EMD-M · NTU and TWITTER | accuracy | Hurt |
| Alternate updates of metric A and flow F (ablation) | EMD-M · NTU and TWITTER | objective value | Improved |
Closest papers by meaning
- DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learning (TPAMI 2022 · similarity 0.88)
- DeepEMD: Few-Shot Image Classification With Differentiable Earth Mover's Distance and Structured Classifiers (CVPR 2020 · similarity 0.87)
- On Computing Diverse Solutions in the Earth Movers Distance (NeurIPS 2026 · similarity 0.85)
- Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport (ICLR 2025 · similarity 0.82)
- Learning Distances from Data with Normalizing Flows and Score Matching (ICML 2025 · similarity 0.82)
- GeoMM: On Geodesic Perspective for Multi-modal Learning (CVPR 2025 · similarity 0.81)
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.
