Research map / Learning theory and algorithms / Fairness
Assessing Fairness in the Presence of Missing Data
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 true inverse probability weights instead of unweighted complete cases | Complete-case fairness estimator · COMPAS MAR | fairness estimation bias: 1.52 -> 1.14 | Helped |
| Use true inverse probability weights instead of unweighted complete cases | Complete-case fairness estimator · COMPAS MNAR | fairness estimation bias: 5.32 -> 3.83 | Helped |
| Use true inverse probability weights instead of unweighted complete cases | Complete-case fairness estimator · ADNI MCAR | fairness estimation bias: 2.90 -> 2.98 | No clear effect |
| Use true inverse probability weights instead of unweighted complete cases | Complete-case fairness estimator · ADNI MAR | fairness estimation bias: 3.80 -> 3.66 | No clear effect |
| Estimate weights with correctly specified logistic regression rather than use unweighted cases | Complete-case fairness estimator · COMPAS MAR | fairness estimation bias: 1.52 -> 1.24 | Helped |
| Estimate weights with a misspecified SVM propensity model rather than use unweighted cases | Complete-case fairness estimator · COMPAS MAR | fairness estimation bias: 1.52 -> 46.5 | Hurt |
| Use working propensity score models instead of true weights | Complete-case fairness estimator · COMPAS MNAR | fairness estimation bias: 3.83 -> 5.11; 3.83 -> 5.32; 3.83 -> 9.64; 3.83 -> 5.05 | Hurt |
| Use uniform weights rather than estimated propensity weights | Complete-case fairness estimator · MCAR | fairness estimation bias | No clear effect |
| Use a more expressive random forest or XGB propensity estimator rather than a correctly specified logistic model | Complete-case fairness estimator · Synthetic regression, MAR | fairness estimation bias | Hurt |
| Increase imbalance between sensitive groups | Fairness estimator using complete cases · Synthetic regression and COMPAS and ADNI | fairness estimation bias | Hurt |
| Increase disparity in feature distributions between sensitive groups | Fairness estimator using complete cases · Synthetic regression | fairness estimation bias | Hurt |
| Use the true propensity weights | Complete-case fairness estimator · Complete data domain, under the theorem's assumptions | fairness estimation error | Helped |
Closest papers by meaning
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.85)
- Who’s the (Multi-)Fairest of Them All: Rethinking Interpolation-Based Data Augmentation Through the Lens of Multicalibration (AAAI 2025 · similarity 0.85)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · similarity 0.85)
- Achievable Fairness on Your Data With Utility Guarantees (NeurIPS 2024 · similarity 0.85)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 2024 · similarity 0.84)
- Multiaccuracy and Multicalibration via Proxy Groups (ICML 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.