Research map / Label-efficient and robust learning / Recognition and long-tail
Relative Order Analysis and Optimization for Unsupervised Deep Metric Learning
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 |
|---|---|---|---|
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · CUB-200-2011 | R@1: +8.6% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · CUB-200-2011 | R@2: +8.3% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · CUB-200-2011 | R@4: +6.2% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · CUB-200-2011 | R@8: +3.5% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · Cars-196 | R@1: +0.0% | No clear effect |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · Cars-196 | R@2: +0.7% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · Cars-196 | R@4: +1.7% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · Cars-196 | R@8: +2.0% | Helped |
| use relative-order prediction and cooperative metric learning | unsupervised metric learning · Stanford Online Products | R@1, R@10, R@100: +2.3%, +2.3%, +1.9% | Helped |
| use relative-order prediction and cooperative metric learning from random initialization | unsupervised metric learning · Stanford Online Products (from scratch) | R@1, R@10, R@100: 3.1%, 2.8%, and 2.6% gain | Helped |
| add the relative order constraint (ablation) | ROUL · CUB-200-2011 | R@1, R@2, R@4, R@8: 55.4 -> 56.1; 67.0 -> 67.7; 76.7 -> 78.1; 84.9 -> 85.9 | Helped |
| add the metric order constraint (ablation) | ROUL with relative order constraint · CUB-200-2011 | R@1, R@2, R@4, R@8: 56.1 -> 56.7; 67.7 -> 68.4; 78.1 -> 78.3; 85.9 -> 86.3 | Helped |
Closest papers by meaning
- Deep Unsupervised Similarity Learning Using Partially Ordered Sets (CVPR 2017 · similarity 0.87)
- Local Similarity-Aware Deep Feature Embedding (NeurIPS 2016 · similarity 0.85)
- Energy Confused Adversarial Metric Learning for Zero-Shot Image Retrieval and Clustering (AAAI 2019 · similarity 0.85)
- Unsupervised Hyperbolic Metric Learning (CVPR 2021 · similarity 0.85)
- Unsupervised Feature Learning via Non-Parametric Instance Discrimination (CVPR 2018 · similarity 0.84)
- Unsupervised Metric Learning with Synthetic Examples (AAAI 2020 · similarity 0.84)
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.