Research map / Label-efficient and robust learning / Semi-supervised learning
Semi-Supervised AUC Optimization Without Guessing Labels of Unlabeled Data
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 |
|---|---|---|---|
| add unlabeled data treated as both positive and negative (SAMULT) to supervised AUC optimization | supervised linear AUC optimization · australian | AUC: .879 -> .903 | Helped |
| add unlabeled data treated as both positive and negative (SAMULT) | supervised linear AUC optimization · clean1 | AUC: .760 -> .777 | Helped |
| add unlabeled data (SAMULT) | supervised linear AUC optimization · madelon | AUC: .510 -> .517 | No clear effect |
| random partition of unlabeled data into pseudo-positive/negative with ensemble of 20 (SAMPURA over SAMULT) (ablation) | SAMULT · credit-g | AUC: .748 -> .761 | Helped |
| ensemble SAMPURA over SAMULT (ablation) | SAMULT · ijcnn1 | AUC: .913 -> .915 | Helped |
| ensemble SAMPURA over SAMULT (ablation) | SAMULT · australian | AUC: .903 -> .903 | No clear effect |
| SAMULT instead of PNU-AUC (needs ground-truth class prior) | semi-supervised AUC · australian | AUC: .903 -> .903 | Helped |
| use SAMULT instead of SSRankBoost | semi-supervised AUC · colic | AUC: .721 -> .869 | Helped |
| use SAMULT P+U (unlabeled as negative) instead of PU-RSVM | positive-unlabeled AUC · australian | AUC: .844 -> .900 | Helped |
| use SAMULT P+U instead of PU-AUC (no prior estimation) | positive-unlabeled AUC · clean1 | AUC: .786 -> .796 | Helped |
| use SAMULT P+U instead of PU-AUC | positive-unlabeled AUC · madelon | AUC: .533 -> .514 | Hurt |
| increase training set size | SAMULT and SAMULT P+U · australian, clean1, wdbc | AUC and cosine similarity to supervised solution | Helped |
Closest papers by meaning
- AUC Optimization from Multiple Unlabeled Datasets (AAAI 2024 · similarity 0.91)
- Semi-Supervised Sparse Gaussian Classification: Provable Benefits of Unlabeled Data (NeurIPS 2024 · similarity 0.86)
- Bi-Level Optimization for Semi-Supervised Learning with Pseudo-Labeling (AAAI 2025 · similarity 0.86)
- High-Dimensional Bayesian Optimization via Semi-Supervised Learning with Optimized Unlabeled Data Sampling (ICML 2024 · similarity 0.86)
- Pairwise AUC Optimization Needs Corrective Power: A Unified View (NeurIPS 2026 · similarity 0.86)
- Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning (NeurIPS 2025 · 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.