Research map / Learning theory and algorithms / Prediction and decision losses
Positive-Unlabeled Learning with Non-Negative Risk Estimator
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 |
|---|---|---|---|
| replace the uPU risk estimator with the non-negative nnPU risk estimator | deep neural network PU learning · MNIST | test risk | Helped |
| replace the uPU risk estimator with the non-negative nnPU risk estimator | deep neural network PU learning · epsilon | test risk | Helped |
| replace the uPU risk estimator with the non-negative nnPU risk estimator | deep neural network PU learning · 20News | test risk | Helped |
| replace the uPU risk estimator with the non-negative nnPU risk estimator | deep neural network PU learning · CIFAR-10 | test risk | Helped |
| overestimate the class prior using πp' = 1.1πp (ablation) | nnPU · MNIST | test risk | Mixed |
| underestimate the class prior (ablation) | nnPU · MNIST | test risk | Hurt |
| overestimate the class prior using πp' = 1.1πp (ablation) | nnPU · epsilon | test risk | Helped |
| overestimate the class prior using πp' = 1.2πp (ablation) | nnPU · epsilon | test risk | No clear effect |
| overestimate the class prior using πp' = 1.1πp (ablation) | nnPU · 20News | test risk | Helped |
| overestimate the class prior using πp' = 1.2πp (ablation) | nnPU · 20News | test risk | Hurt |
| overestimate the class prior using πp' = 1.1πp (ablation) | nnPU · CIFAR-10 | test risk | No clear effect |
| overestimate the class prior using πp' = 1.2πp (ablation) | nnPU · CIFAR-10 | test risk | Helped |
Closest papers by meaning
- Learning from Positive and Unlabeled Data with Arbitrary Positive Shift (NeurIPS 2020 · similarity 0.94)
- Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning (NeurIPS 2025 · similarity 0.94)
- Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms (ICLR 2026 · similarity 0.91)
- Theoretical Comparisons of Positive-Unlabeled Learning against Positive-Negative Learning (NeurIPS 2016 · similarity 0.91)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.90)
- Positive and Unlabeled Learning with Controlled Probability Boundary Fence (ICML 2024 · 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.