Research map / Learning theory and algorithms / Fairness
Achieving Fairness at No Utility Cost via Data Reweighing with Influence
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 influence-based individual sample reweighing | Logistic Regression and Neural Networks · Adult, Compas, Communities and Crime, German Credit | Equal Opportunity fairness gap | Helped |
| Use influence-based individual sample reweighing | Logistic Regression and Neural Networks · Adult, Compas, Communities and Crime, German Credit | predictive accuracy | No clear effect |
| Use influence-based individual sample reweighing | Logistic Regression and Neural Networks · Demographic Parity settings | predictive accuracy | Mixed |
| Use influence-based individual sample reweighing | Logistic Regression and Neural Networks · Demographic Parity | fairness-utility tradeoff | Mixed |
| Replace group-level heuristic reweighing with sample-specific reweighing | Massaging, Reweigh, and Preferential · Adult and Compas | Equal Opportunity fairness | Helped |
| Use Massaging, Reweigh, or Preferential instead of the proposed method | Fair classification · Adult and Compas | Equal Opportunity fairness | Mixed |
| Use Disparate Impact Remover or Label Bias instead of the proposed method | Fair classification · Adult and Compas | predictive accuracy | Hurt |
| Use Adversarial Debiasing or Conditional Adversarial Debiasing instead of the proposed method | Neural Networks · Adult and Compas | predictive accuracy | Mixed |
| Use adversarial debiasing | Neural Networks · Adult and Compas | predictive accuracy | Mixed |
| Tune fairness and utility constraints with validation-set hyperparameters (ablation) | Influence-based reweighing · Adult | fairness and predictive accuracy | Helped |
| Downweight samples rather than simultaneously upweighting and downweighting (ablation) | Influence-based reweighing · Multiple tabular datasets | fairness and predictive accuracy | No clear effect |
| Use group removal rather than individual-sample removal to estimate influence (ablation) | Influence-function estimates · Adult | influence-estimation accuracy | Hurt |
Closest papers by meaning
- Fairness with Adaptive Weights (ICML 2022 · similarity 0.89)
- Self-Supervised Fair Representation Learning without Demographics (NeurIPS 2022 · similarity 0.88)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · similarity 0.88)
- Understanding Fairness and Prediction Error through Subspace Decomposition and Influence Analysis (NeurIPS 2025 · similarity 0.87)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · similarity 0.86)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.86)
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.