Research map / Label-efficient and robust learning / Recognition and long-tail
Recovering True Classifier Performance in Positive-Unlabeled Learning
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 |
|---|---|---|---|
| Correct PU performance estimates using class priors and noise fraction | PU classifier evaluation · 12 UCI datasets | AUC mean absolute error | Helped |
| Correct the AUC-PR curve indirectly using class priors and noise fraction | PU classifier evaluation · 12 UCI datasets | AUC-PR mean absolute error | Helped |
| Use indirect rather than direct recovery (ablation) | AUC recovery · 12 UCI datasets | AUC estimation accuracy: P = 6.5 · 10⁻⁶; P = 5.7 · 10⁻³ | Helped |
| Estimate α and β with AlphaMax instead of using their real values (ablation) | Indirect and direct recovery · 12 UCI datasets | AUC estimation accuracy | No clear effect |
| Use uncorrected PU estimates as noise increases | PU classifier evaluation · 12 UCI datasets | AUC estimation accuracy | Hurt |
| Estimate α and β with larger absolute error (ablation) | Estimated-parameter recovery · 12 UCI datasets | AUC estimation accuracy | Hurt |
| Use estimated rather than real α and β (ablation) | Performance recovery · 12 UCI datasets | AUC-PR estimation accuracy | Mixed |
Closest papers by meaning
- Class Prior Estimation with Biased Positives and Unlabeled Examples (AAAI 2020 · similarity 0.93)
- Learning from Positive and Unlabeled Data with Arbitrary Positive Shift (NeurIPS 2020 · similarity 0.91)
- Estimating Accuracy from Unlabeled Data: A Bayesian Approach (ICML 2016 · similarity 0.90)
- Estimating the class prior and posterior from noisy positives and unlabeled data (NeurIPS 2016 · similarity 0.90)
- Rethinking Class-Prior Estimation for Positive-Unlabeled Learning (ICLR 2022 · similarity 0.89)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.89)
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.