Research map / Learning theory and algorithms / Prediction and decision losses
Does It Pay to Optimize AUC?
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 |
|---|---|---|---|
| exactly optimize linear AUC with AUC-opt rather than use B-LR | linear classifiers on t-SNE datasets · 50 t-SNE datasets, R² training | datasets with significantly higher AUC: 17 | Helped |
| exactly optimize linear AUC with AUC-opt rather than use B-LR | linear classifiers on t-SNE datasets · 50 t-SNE datasets, R² testing | datasets with significantly higher AUC: 11 | Helped |
| exactly optimize linear AUC with AUC-opt rather than use B-LR | linear classifiers on t-SNE datasets · 50 t-SNE datasets, R³ training | datasets with significantly higher AUC: 4 | Helped |
| exactly optimize linear AUC with AUC-opt rather than use B-LR | linear classifiers on t-SNE datasets · 50 t-SNE datasets, R³ testing | datasets with significantly higher AUC: 2 | No clear effect |
| add class weighting | SVM and LR · 50 real-world datasets | AUC | Helped |
| add class weighting | nonlinear classifiers · 50 real-world datasets | AUC | No clear effect |
| use approximate nonlinear AUC optimizers instead of popular nonlinear standard classifiers | nonlinear classifiers · 50 real-world datasets, testing | AUC | Hurt |
Closest papers by meaning
- Interplay of ROC and Precision-Recall AUCs: Theoretical Limits and Practical Implications in Binary Classification (ICML 2024 · similarity 0.89)
- AUC Optimization from Multiple Unlabeled Datasets (AAAI 2024 · similarity 0.87)
- Positive-unlabeled AUC Maximization under Covariate Shift (ICML 2025 · similarity 0.87)
- AUC Maximization under Positive Distribution Shift (NeurIPS 2024 · similarity 0.86)
- A Closer Look at AUROC and AUPRC under Class Imbalance (NeurIPS 2024 · similarity 0.85)
- Pairwise AUC Optimization Needs Corrective Power: A Unified View (NeurIPS 2026 · similarity 0.84)
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.