Research map / Learning theory and algorithms / Fairness
Accurate Fairness: Improving Individual Fairness without Trading Accuracy
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 |
|---|---|---|---|
| Train with Siamese fairness using similar individuals as parallel inputs and accurate-fairness constraints | FCNN classifiers · Adult, German Credit, COMPAS | ACC: 0.105 ACC (13.69% higher ACC) | Helped |
| Train with Siamese fairness using similar individuals as parallel inputs and accurate-fairness constraints | FCNN classifiers · Adult, German Credit, COMPAS | FTA: 0.080 FTA (8.78% higher FTA) | Helped |
| Train with Siamese fairness using similar individuals as parallel inputs and accurate-fairness constraints | FCNN classifiers · Adult, German Credit, COMPAS | TFR: 0.148 TFR (20.57% higher TFR) | Helped |
| Train with Siamese fairness using similar individuals as parallel inputs and accurate-fairness constraints | FCNN classifiers · Adult, German Credit, COMPAS | F-F1: 0.085 F-F1 score (10.01% higher F-F1 score) | Helped |
| Use only the maximum and minimum sensitive-attribute values for augmentation instead of all values (ablation) | Siamese Fairness (SF) · Adult, German Credit, COMPAS (FCNNs) | F-F1: 0.930±0.003 -> 0.929±0.004 | No clear effect |
| Use only the maximum and minimum sensitive-attribute values for augmentation (ablation) | Siamese Fairness (SF) · Ctrip | TFR: 0.412±0.068 -> 0.620±0.013; 0.281±0.145 -> 0.637±0.010 | Helped |
| Apply Siamese fairness in-processing | Logistic regression classifiers · Adult, German Credit, COMPAS | ACC: 0.072 ACC (9.03% higher ACC) | Helped |
| Apply Siamese fairness in-processing | Logistic regression classifiers · Adult, German Credit, COMPAS | FTA: 0.01 FTA (1.02% higher FTA) | No clear effect |
| Apply Siamese fairness in-processing | Logistic regression classifiers · Adult, German Credit, COMPAS | F-F1: 0.048 F-F1 score (5.43% higher F-F1 score) | Helped |
| Apply Siamese fairness in-processing | SVM classifiers · Adult, German Credit, COMPAS | ACC: 0.063 ACC (8.38% higher ACC) | Mixed |
| Apply Siamese fairness in-processing | SVM classifiers · Adult, German Credit, COMPAS | FTA: 0.045 FTA (4.84% higher FTA) | Helped |
| Apply Siamese fairness in-processing | SVM classifiers · Adult, German Credit, COMPAS | F-F1: 0.052 F-F1 score (6.22% higher F-F1 score) | Helped |
Closest papers by meaning
- Achievable Fairness on Your Data With Utility Guarantees (NeurIPS 2024 · similarity 0.90)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · similarity 0.89)
- FairNet: Dynamic Fairness Correction without Performance Loss via Contrastive Conditional LoRA (NeurIPS 2025 · similarity 0.89)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.88)
- FFB: A Fair Fairness Benchmark for In-Processing Group Fairness Methods (ICLR 2024 · similarity 0.88)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 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.