Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Variational Approach for Learning from Positive and Unlabeled Data
Hui Chen, Fangqing Liu, Yin Wang, Liyue Zhao, Hao Wu
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 MixUp consistency regularization (ablation) | VPU · FashionMNIST 1,4,7 positive, 1000 labeled | accuracy (%): 87.2 -> 91.3 | Improved |
| MixUp between P and U instead of P only (ablation) | VPU · FashionMNIST 1,4,7 positive, 1000 labeled | accuracy (%): 90.3 -> 91.3 | Improved |
| MixUp between P and U instead of among P union U (ablation) | VPU · FashionMNIST 1,4,7 positive, 1000 labeled | accuracy (%): 90.1 -> 91.3 | Improved |
| use mean squared logarithmic error instead of MSE for consistency (ablation) | VPU · FashionMNIST 1,4,7 positive, 1000 labeled | accuracy (%): 90.0 -> 91.3 | Improved |
| use VPU instead of nnPU (estimated class prior) | nnPU · CIFAR-10 1 | accuracy (%): 85.6 -> 89.5 | Improved |
| use VPU instead of nnPU (estimated class prior) | nnPU · Grid Stability 1 | accuracy (%): 80.8 -> 92.6 | Improved |
| use VPU instead of nnPU | nnPU · Page Blocks 1 | accuracy (%): 93.4 -> 93.6 | No clear effect |
| use VPU instead of uPU | uPU · Grid Stability 1 | accuracy (%): 92.6 -> 92.6 | No clear effect |
| use VPU instead of RP | RP · FashionMNIST 1 | accuracy (%): 92.2 -> 92.7 | No clear effect |
| use VPU instead of RP | RP · STL-10 2 | accuracy (%): 68.5 -> 83.7 | Improved |
| use VPU instead of GenPU | GenPU · FashionMNIST 1 | accuracy (%): 47.8 -> 92.7 | Improved |
Closest papers by meaning
- Learning from Positive and Unlabeled Data with Arbitrary Positive Shift (NeurIPS 2020 · similarity 0.89)
- Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning (NeurIPS 2025 · similarity 0.89)
- Rethinking Class-Prior Estimation for Positive-Unlabeled Learning (ICLR 2022 · similarity 0.89)
- Positive-Unlabeled Learning with Non-Negative Risk Estimator (NeurIPS 2017 · similarity 0.88)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.87)
- 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.
