Research map / Label-efficient and robust learning / Recognition and long-tail
Fast Nonparametric Estimation of Class Proportions in the Positive-Unlabeled Classification Setting
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 |
|---|---|---|---|
| use distance-curve estimator with learned neural net vs best prior estimators | class prior estimation · 40 PU datasets | Mean Rank Error: 1.7 vs 2.8 (next best) | Helped |
| apply a univariate nontraditional-classifier transform before distance curve construction (ablation) | Distance algorithm · multi-dimensional datasets | MAE / MRE | Helped |
| combine all six transforms into a six-dimensional vector instead of one optimal transform (ablation) | Distance algorithm · multi-dimensional datasets | MAE / MRE | Helped |
| vary distance function (Minkowski p=1, p=2, Yang's distance p=1, p=2) (ablation) | Distance algorithm on untransformed data · multi-dimensional datasets | MAE | No clear effect |
| reduce neural network size to a few dozen hidden neurons (ablation) | Distance algorithm with transforms · transformed data | performance | No clear effect |
| use larger neural network (ablation) | Distance algorithm on original data · untransformed data | performance | Helped |
| Distance algorithm vs Ramaswamy et al. on running time | class prior estimation · multi-dimensional datasets | mean time (sec): 2818.1 -> 66.4 | Helped |
| Distance algorithm on datasets with poor classifier separation (Abalone) | class prior estimation · Abalone | MAE: 0.380 vs 0.316 (Elkan & Noto) | Hurt |
Closest papers by meaning
- Class Prior Estimation with Biased Positives and Unlabeled Examples (AAAI 2020 · similarity 0.91)
- Rethinking Learning from Label Proportions via Moment Matching (NeurIPS 2026 · similarity 0.86)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.86)
- Class-Attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective (AAAI 2024 · similarity 0.86)
- Balanced Data, Imbalanced Spectra: Unveiling Class Disparities with Spectral Imbalance (ICML 2024 · similarity 0.85)
- Learning from Label Proportions: Bootstrapping Supervised Learners via Belief Propagation (ICLR 2024 · similarity 0.85)
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.