Research map / Learning theory and algorithms / Fairness
Fair Classification with Adversarial Perturbations
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 Err-Tol instead of CHKV under adversarial protected-attribute perturbations | Synthetic data classifier · Synthetic data, η=3% and η=5% | accuracy: 0.859 (.143) -> 1.00 (.001); 0.799 (.139) -> .999 (.002) | Helped |
| Use Err-Tol instead of CHKV under adversarial protected-attribute perturbations | Synthetic data classifier · Synthetic data, η=3% and η=5% | statistical rate: .787 (.015) -> .799 (.002); .795 (.049) -> .799 (.004) | Mixed |
| Apply Err-Tol rather than the unconstrained classifier | COMPAS classifier · COMPAS, A_TN, η=3.5% | statistical rate: 0.80 -> 0.92 | Helped |
| Apply Err-Tol rather than the unconstrained classifier | COMPAS classifier · COMPAS, A_TN, η=3.5% | accuracy: 0.67 -> 0.60 | Hurt |
| Apply Err-Tol rather than AKM | COMPAS classifier · COMPAS, A_TN, η=3.5% | statistical rate: 0.87 -> 0.92 | Helped |
| Apply Err-Tol rather than the unconstrained classifier | COMPAS classifier · COMPAS, A_FN, η=3.5% | statistical rate: 0.80 -> 0.91 | Helped |
| Apply Err-Tol rather than the unconstrained classifier | COMPAS classifier · COMPAS, A_FN, η=3.5% | accuracy: 0.67 -> 0.61 | Hurt |
| Apply Err-Tol rather than AKM | COMPAS classifier · COMPAS, A_FN, η=3.5% | statistical rate: 0.83 -> 0.91 | Helped |
| Apply Err-Tol rather than AKM | COMPAS classifier · COMPAS, A_FN, η=3.5% | accuracy: 0.58 -> 0.61 | Helped |
| Apply Err-Tol rather than LMZV | COMPAS classifier · COMPAS, A_FN, η=3.5% | statistical rate: 0.97 -> 0.91 | Hurt |
Closest papers by meaning
- On the Alignment between Fairness and Accuracy: from the Perspective of Adversarial Robustness (ICML 2025 · similarity 0.91)
- Simple and Effective Specialized Representations for Fair Classifiers (NeurIPS 2025 · similarity 0.89)
- On the Fairness ROAD: Robust Optimization for Adversarial Debiasing (ICLR 2024 · similarity 0.89)
- Can Private Machine Learning Be Fair? (AAAI 2025 · similarity 0.88)
- Optimal Fair Learning Robust to Adversarial Distribution Shift (ICML 2025 · similarity 0.88)
- Adversarial Fairness Network (AAAI 2024 · similarity 0.88)
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.