Research map / Learning theory and algorithms / Fairness
Causal Context Connects Counterfactual Fairness to Robust Prediction and Group Fairness
Jacy Reese Anthis, Victor Veitch
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Counterfactually fair (CF) vs naive predictor on target accuracy, measurement error | counterfactually fair predictor · Measurement Error | target accuracy: 0.8031 -> 0.8204 | Improved |
| CF vs naive on target accuracy, selection on label | counterfactually fair predictor · Selection on Label | target accuracy: 0.8566 -> 0.8663 | Improved |
| CF vs naive on target accuracy, selection on predictors | counterfactually fair predictor · Selection on Predictors | target accuracy: 0.8700 -> 0.8699 | No clear effect |
| CF vs FTU on target accuracy, measurement error | counterfactually fair predictor · Measurement Error | target accuracy: 0.8114 -> 0.8204 | Improved |
| CF vs naive on source accuracy, measurement error | counterfactually fair predictor · Measurement Error | source accuracy: 0.8283 -> 0.7695 | Hurt |
| CF vs naive on source accuracy, selection on label | counterfactually fair predictor · Selection on Label | source accuracy: 0.8771 -> 0.8610 | Hurt |
| CF target accuracy vs predictor trained directly on target, selection on label | counterfactually fair predictor · Selection on Label | target accuracy: 0.8663 -> 0.8655 | No clear effect |
| CF demographic parity difference under measurement error | counterfactually fair predictor · Measurement Error | demographic parity difference: -0.0005 | Improved |
| CF equalized odds difference under selection on label | counterfactually fair predictor · Selection on Label | equalized odds difference: -0.0021 | Improved |
| CF calibration difference under selection on predictors | counterfactually fair predictor · Selection on Predictors | calibration difference: 0.0040 | Improved |
| CF calibration difference under measurement error (mismatched metric) | counterfactually fair predictor · Measurement Error | calibration difference: -0.8158 | Hurt |
Closest papers by meaning
- Fairness-Accuracy Trade-Offs: A Causal Perspective (AAAI 2025 · similarity 0.91)
- Counterfactual Fairness by Combining Factual and Counterfactual Predictions (NeurIPS 2024 · similarity 0.90)
- Fairness on Principal Stratum: A New Perspective on Counterfactual Fairness (ICML 2025 · similarity 0.90)
- Towards counterfactual fairness through auxiliary variables (ICLR 2025 · similarity 0.89)
- FairPFN: A Tabular Foundation Model for Causal Fairness (ICML 2025 · similarity 0.89)
- Understanding challenges to the interpretation of disaggregated evaluations of algorithmic fairness (NeurIPS 2025 · similarity 0.89)
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.
