Research map / Label-efficient and robust learning / Recognition and long-tail
Multi-Label Ranking From Positive and Unlabeled Data
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 appropriately weighted loss and a symmetric surrogate loss | Multi-label PU ranking · Synthetic dataset, Experiment A | mean average precision (samples) | Helped |
| Add appropriately weighted loss (ablation) | Multi-label PU ranking · Synthetic dataset, Experiment A | mean average precision (samples) | Helped |
| Add a symmetric surrogate loss, using ramp loss rather than hinge loss (ablation) | Multi-label PU ranking · Synthetic dataset, Experiment A | mean average precision (samples) | Helped |
| Estimate class priors from unlabeled data | Method 3 · Synthetic dataset, Experiment B | mean average precision (samples) | Mixed |
| Use estimated rather than optimal class priors (ablation) | Method 3 · Synthetic dataset, Experiment B, increasing feature dimensions | mean average precision (samples) | Helped |
| Use Method 3 with estimated class priors instead of the baseline | Multi-label PU ranking · MSCOCO | mean average precision (samples) | Helped |
| Use Method 3 with estimated class priors instead of the baseline | Multi-label PU ranking · NUS-WIDE | mean average precision (samples) | Helped |
| Use a symmetric surrogate loss rather than a non-symmetric hinge loss (ablation) | Weighted multi-label PU ranking · Synthetic dataset with many labels per sample | mean average precision (samples) | Helped |
Closest papers by meaning
- Dist-PU: Positive-Unlabeled Learning From a Label Distribution Perspective (CVPR 2022 · similarity 0.89)
- Multi-Label Learning From Single Positive Labels (CVPR 2021 · similarity 0.89)
- Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms (ICLR 2026 · similarity 0.88)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.87)
- Addressing Multi-Label Learning with Partial Labels: From Sample Selection to Label Selection (AAAI 2025 · similarity 0.86)
- Active Learning From Positive and Unlabeled Examples (NeurIPS 2026 · similarity 0.86)
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.