Research map / Label-efficient and robust learning / Semi-supervised learning
Dist-PU: Positive-Unlabeled Learning From a Label Distribution Perspective
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 |
|---|---|---|---|
| Replace the strongest listed baseline with Dist-PU | PU learning · F-MNIST | ACC (%): 94.82 (0.16) -> 95.40 (0.34) | Helped |
| Replace the strongest listed baseline with Dist-PU | PU learning · CIFAR-10 | ACC (%): 89.83 (0.30) -> 91.88 (0.52) | Helped |
| Replace the strongest listed baseline with Dist-PU | PU learning · Alzheimer | ACC (%): 70.88 (0.72) -> 71.57 (0.62) | Helped |
| Add label distribution alignment (ablation) | Dist-PU · CIFAR-10 | ACC (%): 84.92 (0.89) -> 89.30 (0.48) | Helped |
| Add label distribution alignment (ablation) | Dist-PU with Mixup · CIFAR-10 | ACC (%): 86.97 (0.23) -> 91.62 (0.47) | Helped |
| Add label distribution alignment (ablation) | Dist-PU with entropy minimization and Mixup · CIFAR-10 | ACC (%): 87.67 (0.52) -> 91.88 (0.52) | Helped |
| Add entropy minimization (ablation) | Dist-PU with label distribution alignment · CIFAR-10 | ACC (%): 89.30 (0.48) -> 89.41 (0.48) | Helped |
| Add entropy minimization (ablation) | Dist-PU with label distribution alignment and Mixup · CIFAR-10 | ACC (%): 91.62 (0.47) -> 91.88 (0.52) | Helped |
| Add Mixup (ablation) | Dist-PU with label distribution alignment · CIFAR-10 | ACC (%): 89.30 (0.48) -> 91.62 (0.47) | Helped |
| Add Mixup (ablation) | Dist-PU with label distribution alignment and entropy minimization · CIFAR-10 | ACC (%): 89.41 (0.48) -> 91.88 (0.52) | Helped |
| Add Mixup (ablation) | Entropy-minimization-only variant · CIFAR-10 | ACC (%): 87.50 (0.24) -> 87.67 (0.52) | No clear effect |
Closest papers by meaning
- A Closer Look to Positive-Unlabeled Learning from Fine-grained Perspectives: An Empirical Study (NeurIPS 2025 · similarity 0.93)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.93)
- Positive and Unlabeled Learning with Controlled Probability Boundary Fence (ICML 2024 · similarity 0.89)
- Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical (ICML 2024 · similarity 0.89)
- Active Learning From Positive and Unlabeled Examples (NeurIPS 2026 · similarity 0.88)
- Predicting Label Distribution from Ternary Labels (NeurIPS 2024 · similarity 0.87)
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.