Research map / Label-efficient and robust learning / Recognition and long-tail
Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a balanced-error risk estimator for positive-unlabeled learning that aims to equalize positive and negative classification error rates, based only on the abstract's framing against nnPU.
Problem
Positive-unlabeled (PU) learning, where only positive and unlabeled data are available, and existing risk estimators such as nnPU that estimate the negative-class risk in a way the abstract suggests is problematic.
Approach
Rather than estimating the negative-class risk as in nnPU, the work targets balancing the positive and negative classification error rates within the PU risk estimator; the full method details are not available in the provided text.
Experimental setup
Baselines: nnPU
Builds on: nnPU
Closest papers by meaning
- Positive-Unlabeled Learning with Non-Negative Risk Estimator (NeurIPS 2017 · similarity 0.94)
- Learning from Positive and Unlabeled Data with Arbitrary Positive Shift (NeurIPS 2020 · similarity 0.93)
- Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms (ICLR 2026 · similarity 0.93)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.91)
- Positive and Unlabeled Learning with Controlled Probability Boundary Fence (ICML 2024 · similarity 0.90)
- Active Learning From Positive and Unlabeled Examples (NeurIPS 2026 · 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.