Research map / Learning theory and algorithms / Prediction and decision losses
When False Positive is Intolerant: End-to-End Optimization with Low FPR for Multipartite Ranking
Peisong Wen, Qianqian Xu, Zhiyong Yang, Yuan He, Qingming Huang
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 |
|---|---|---|---|
| CBA-CR instead of cross-entropy training | ReXNet200 · DR TPR@FPR=0.003 | TPR (%): 35.2 -> 39.1 | Improved |
| CBA-CR instead of cross-entropy training | ReXNet200 · UTKFace TPR@FPR=0.003 | TPR (%): 33.1 -> 44.9 | Improved |
| CBA-CR instead of cross-entropy training | ReXNet200 · AgePrediction TPR@FPR=0.003 | TPR (%): 11.3 -> 16.0 | Improved |
| CBA-CR instead of Eban's method | ReXNet200 · NSFW TPR@FPR=0.003 | TPR (%): 56.3 -> 58.1 | Improved |
| larger FPR budget reduces gain over best competitor on DR | CBA-CR · DR | TPR gain over best competitor: +2.3, +2.4, +0.6 at alpha 0.003, 0.01, 0.03 | Mixed |
| existing ranking/ordinal methods versus CE baseline | NNRank, F&H, SoftLabel, DDAG · DR/UTKFace/NSFW/AgePrediction | TPR@FPR | No clear effect |
| CBA-CR at lower FPR gives larger improvements | CBA-CR · all datasets | TPR gain: UTKFace 33.1 -> 44.9 at 0.003 vs 80.9 -> 82.9 at 0.03 | Improved |
| CBA-CR versus CE on AUC | ReXNet200 · DR AUC | AUC (%): 84.4 -> 86.3 | Improved |
| CBA-CR versus SoftLabel on AUC | ReXNet200 · UTKFace AUC | AUC (%): 96.0 -> 95.9 | No clear effect |
| add AUC loss to F&H | F&H · UTKFace TPR@FPR=0.003 | TPR (%): 37.5 -> 42.1 | Improved |
| cache size Nc too small or too large (ablation) | CBA-CR · DR sweep Nc in {64..2048} | TPR@FPR | Hurt |
| cache size effect at alpha 0.003 (ablation) | CBA-CR · DR | TPR@FPR=0.003 | No clear effect |
Closest papers by meaning
- Large-scale Optimization of Partial AUC in a Range of False Positive Rates (NeurIPS 2022 · similarity 0.86)
- When All We Need is a Piece of the Pie: A Generic Framework for Optimizing Two-way Partial AUC (ICML 2021 · similarity 0.85)
- Optimizing Two-Way Partial AUC With an End-to-End Framework (TPAMI 2022 · similarity 0.85)
- StochasticRank: Global Optimization of Scale-Free Discrete Functions (ICML 2020 · similarity 0.84)
- Pairwise AUC Optimization Needs Corrective Power: A Unified View (NeurIPS 2026 · similarity 0.83)
- Offline Model-Based Optimization by Learning to Rank (ICLR 2025 · 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.
