Research map / Learning theory and algorithms / Prediction and decision losses
AUC Optimization with a Reject Option
Song-Qing Shen, Bin-Bin Yang, Wei Gao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| lower reject cost d from 0.5 to 0.42 in AUCRO (ablation) | AUCRO · diabetes | test AUC: .8318 -> .9129 | Improved |
| lower reject cost d from 0.5 to 0.42 in AUCRO (ablation) | AUCRO · german | test AUC: .7896 -> .8826 | Improved |
| lower reject cost d from 0.5 to 0.42 in AUCRO (ablation) | AUCRO · splice | test AUC: .9243 -> .9810 | Improved |
| lower reject cost d from 0.5 to 0.42 in AUCRO (ablation) | AUCRO · w8a | test AUC: .9703 -> .9954 | Improved |
| lower reject cost d from 0.5 to 0.42 in AUCRO (ablation) | AUCRO · covtype | test AUC: .8232 -> .9155 | Improved |
| AUCRO with no rejection (d=0.5) vs OAMseq | AUCRO · connect-4 | test AUC: .7351 -> .8578 | Improved |
| AUCRO with no rejection (d=0.5) vs OAMgd | AUCRO · covtype | test AUC: .7459 -> .8232 | Improved |
| AUCRO with no rejection (d=0.5) vs OPAUC | AUCRO · diabetes | test AUC: .8335 -> .8318 | No clear effect |
| AUCRO with no rejection (d=0.5) vs OPAUC | AUCRO · german | test AUC: .7911 -> .7896 | No clear effect |
| AUCRO with no rejection (d=0.5) vs FSAUC | AUCRO · w8a | test AUC: .9482 -> .9703 | Improved |
| increase stepsize eta_t above 2^-4 (ablation) | AUCRO (d=0.47) · Figure 2 datasets | AUC | Hurt |
| vary stepsize eta_t within [2^-12, 2^-4] (ablation) | AUCRO (d=0.47) · Figure 2 datasets | AUC | No clear effect |
Closest papers by meaning
- Does It Pay to Optimize AUC? (AAAI 2023 · similarity 0.85)
- Optimal Strategies for Reject Option Classifiers (JMLR 2023 · similarity 0.85)
- Regression with reject option and application to kNN (NeurIPS 2020 · similarity 0.84)
- Closing the Approximation Gap of Partial AUC Optimization: A Tale of Two Formulations (TPAMI 2025 · similarity 0.83)
- Pairwise AUC Optimization Needs Corrective Power: A Unified View (NeurIPS 2026 · similarity 0.83)
- AUC Optimization from Multiple Unlabeled Datasets (AAAI 2024 · similarity 0.82)
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.
