Research map / Learning theory and algorithms / Prediction and decision losses
Margin Based PU Learning
Tieliang Gong, Guangtao Wang, Jieping Ye, Zongben Xu, Ming Lin
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 |
|---|---|---|---|
| add PMPU re-training to Logistic Regression | LR · WAVEFORM 5% | accuracy: 0.6612 -> 0.8033 | Improved |
| add PMPU re-training to Logistic Regression | LR · MNIST 1% | accuracy: 0.7386 -> 0.8455 | Improved |
| add PMPU re-training to RF | RF · MNIST 10% | accuracy: 0.9378 -> 0.9456 | Improved |
| add PMPU re-training to GBT | GBT · CIFAR-100 1% | accuracy: 0.1497 -> 0.1776 | Improved |
| add PMPU re-training to LR | LR · RCV-1 10% | accuracy: 0.7392 -> 0.7784 | Improved |
| PMPU vs LRPU at low sampling ratio | PMPU · CIFAR-10/CIFAR-100 (<5% sampling) | accuracy | Hurt |
| PMPU vs LRPU at sampling ratio above 5% | PMPU · 6 datasets | accuracy | Improved |
| iterative re-training with PMPU at 1% sampling ratio | PMPU · CIFAR-10, CIFAR-100, RCV-1 | accuracy | No clear effect |
| increase labeled positive sampling ratio | PMPU · WAVEFORM | accuracy: 72.99 -> 85.63 | Improved |
Closest papers by meaning
- Loss Decomposition and Centroid Estimation for Positive and Unlabeled Learning (TPAMI 2019 · similarity 0.87)
- Learning from Positive and Unlabeled Data with a Selection Bias (ICLR 2019 · similarity 0.87)
- Efficient Training for Positive Unlabeled Learning (TPAMI 2018 · similarity 0.87)
- Theoretical Comparisons of Positive-Unlabeled Learning against Positive-Negative Learning (NeurIPS 2016 · similarity 0.86)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.86)
- Positive and Unlabeled Learning with Controlled Probability Boundary Fence (ICML 2024 · 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.
