Research map / Learning theory and algorithms / Fairness
Re-weighting Based Group Fairness Regularization via Classwise Robust Optimization
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 FairDRO instead of PL | Logistic regression on Adult · Adult | DCA: 2.55 ($\pm$0.39) -> 1.99 ($\pm$0.38) | Helped |
| Use FairDRO instead of RW | Logistic regression on Adult · Adult | accuracy: 81.98 ($\pm$0.04) -> 79.23 ($\pm$0.11) | Hurt |
| Use FairDRO instead of LBC | Logistic regression on COMPAS · COMPAS | DCA: 6.03 ($\pm$1.13) -> 5.20 ($\pm$0.92) | Helped |
| Use FairDRO instead of Scratch | Logistic regression on COMPAS · COMPAS | accuracy: 63.29 ($\pm$0.00) -> 61.97 ($\pm$0.24) | Hurt |
| Use FairDRO instead of LBC | ResNet18 on UTKFace · UTKFace | DCA | Helped |
| Use FairDRO instead of LBC | BERT on CivilComments · CivilComments | DCA | Mixed |
| Replace classwise DRO with direct cross-entropy approximation of the DCA-regularized objective (ablation) | FairDRO · Adult, COMPAS, UTKFace, and CivilComments | accuracy-DCA trade-off | Hurt |
| Replace classwise DRO with classwise variance regularization (ablation) | FairDRO · Adult, COMPAS, UTKFace, and CivilComments | accuracy-DCA trade-off | Hurt |
| Use one uncertainty set over class-group pairs instead of separate classwise uncertainty sets (ablation) | FairDRO · COMPAS, UTKFace, and CivilComments | accuracy-DCA trade-off | Hurt |
| Use one uncertainty set over class-group pairs instead of separate classwise uncertainty sets (ablation) | FairDRO · Adult | accuracy-DCA trade-off | No clear effect |
Closest papers by meaning
- Enhancing Group Fairness in Online Settings Using Oblique Decision Forests (ICLR 2024 · similarity 0.90)
- FFB: A Fair Fairness Benchmark for In-Processing Group Fairness Methods (ICLR 2024 · similarity 0.88)
- Fairness without Demographics through Shared Latent Space-Based Debiasing (AAAI 2024 · similarity 0.88)
- Metric-Agnostic Continual Learning for Sustainable Group Fairness (AAAI 2025 · similarity 0.88)
- On the Fairness ROAD: Robust Optimization for Adversarial Debiasing (ICLR 2024 · similarity 0.87)
- Towards Harmless Rawlsian Fairness Regardless of Demographic Prior (NeurIPS 2024 · similarity 0.87)
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.