Research map / Learning theory and algorithms / Prediction and decision losses
Optimal Strategies for Reject Option Classifiers
Vojtech Franc, Daniel Prusa, Vaclav Voracek
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 |
|---|---|---|---|
| learn the uncertainty score with SELE loss instead of using the MCP score | Logistic Regression classifier · 11 classification datasets | average rank of AuRC: 2.73 -> 1.36 | Improved |
| learn the uncertainty score with SELE loss instead of the Margin score | SVM classifier · 11 classification datasets | average rank of AuRC: 2.82 -> 1.09 | Improved |
| learn the uncertainty score with SELE loss instead of the Margin score | SVOR classifier · 11 ordinal regression datasets | average rank of AuRC: 3.00 -> 1.27 | Improved |
| use REG score instead of Margin score | SVOR classifier · 11 ordinal regression datasets | average rank of AuRC: 3.00 -> 1.73 | Improved |
| use REG score instead of Margin score | SVM classifier · 11 classification datasets | average rank of AuRC: 2.82 -> 2.09 | Improved |
| use REG score instead of MCP score | Logistic Regression classifier · 11 classification datasets | average rank of AuRC: 2.73 -> 3.55 | Hurt |
| use SELE score instead of TCP score | Logistic Regression classifier · 11 classification datasets | average rank of AuRC: 2.36 -> 1.36 | Improved |
| use SELE score instead of the Margin score | SVM classifier · SENSORLESS | AuRC: 3.68 -> 1.56 | Improved |
| use SELE score instead of the Margin score | SVM classifier · SHUTTLE | AuRC: 1.31 -> 0.24 | Improved |
| use SELE score instead of MCP score | Logistic Regression classifier · PENDIGIT | AuRC: 0.69 -> 1.55 | Hurt |
| use SELE score instead of MCP score | Logistic Regression classifier · COVTYPE | AuRC: 16.49 -> 17.58 | Hurt |
| use SELE score instead of the Margin score | SVOR classifier · MSD | AuRC: 6.23 -> 4.26 | Improved |
Closest papers by meaning
- On discriminative learning of prediction uncertainty (ICML 2019 · similarity 0.93)
- Regression with Cost-based Rejection (NeurIPS 2023 · similarity 0.87)
- Leveraging Uncertainty Estimates To Improve Classifier Performance (ICLR 2024 · similarity 0.85)
- Rejection via Learning Density Ratios (NeurIPS 2024 · similarity 0.85)
- What Does It Take to Build a Performant Selective Classifier? (NeurIPS 2025 · similarity 0.84)
- One-Shot Strategic Classification Under Unknown Costs (ICML 2024 · similarity 0.83)
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.
