Research map / Label-efficient and robust learning / Recognition and long-tail
Class-Weighted Classification: Trade-offs and Robust Approaches
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 |
|---|---|---|---|
| Use LCVaR instead of standard risk | Logistic regression · Covertype test | Worst Class Risk: 0.5111 -> 0.5037 | Helped |
| Use LHCVaR instead of standard risk | Logistic regression · Covertype test | Worst Class Risk: 0.5111 -> 0.4907 | Helped |
| Use LCVaR instead of balanced risk | Logistic regression · Covertype test | Worst Class Risk: 0.5333 -> 0.5037 | Helped |
| Use LHCVaR instead of balanced risk | Logistic regression · Covertype test | Worst Class Risk: 0.5333 -> 0.4907 | Helped |
| Use LCVaR instead of standard risk | Logistic regression · Covertype test | Standard Risk: 0.3275 -> 0.3384 | Hurt |
| Use LHCVaR instead of standard risk | Logistic regression · Covertype test | Standard Risk: 0.3275 -> 0.3979 | Hurt |
| Set LCVaR α to 0.01 rather than 0.05 (ablation) | LCVaR logistic regression · Covertype test | Worst Class Risk: 0.4932 -> 0.5474 | Hurt |
| Set LCVaR α to 0.1 rather than 0.05 (ablation) | LCVaR logistic regression · Covertype test | Worst Class Risk: 0.4932 -> 0.5037 | Hurt |
| Set LHCVaR κ to 0.8 rather than 1 (ablation) | LHCVaR logistic regression · Covertype test | Worst Class Risk: 0.4907 -> 0.5408 | Hurt |
| Set LHCVaR κ to 1.2 rather than 1 (ablation) | LHCVaR logistic regression · Covertype test | Worst Class Risk: 0.4907 -> 0.5050 | Hurt |
Closest papers by meaning
- Class-Attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective (AAAI 2024 · similarity 0.87)
- Label Distributionally Robust Losses for Multi-class Classification: Consistency, Robustness and Adaptivity (ICML 2023 · similarity 0.87)
- Balancing the Scales: A Theoretical and Algorithmic Framework for Learning from Imbalanced Data (ICML 2025 · similarity 0.86)
- Toward Robustness in Multi-Label Classification: A Data Augmentation Strategy against Imbalance and Noise (AAAI 2024 · similarity 0.85)
- Learning Imbalanced Data with Beneficial Label Noise (ICML 2025 · similarity 0.84)
- Concentration and excess risk bounds for imbalanced classification with synthetic oversampling (NeurIPS 2025 · 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.