AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Causal Context Connects Counterfactual Fairness to Robust Prediction and Group Fairness

Jacy Reese Anthis, Victor Veitch

NeurIPS 2023 · Learning theory and algorithms › Fairness · approach: Post-processing and transport for fairness · 2 citations (OpenAlex)

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 changedWhereMeasuredEffect
Counterfactually fair (CF) vs naive predictor on target accuracy, measurement errorcounterfactually fair predictor · Measurement Errortarget accuracy: 0.8031 -> 0.8204Improved
CF vs naive on target accuracy, selection on labelcounterfactually fair predictor · Selection on Labeltarget accuracy: 0.8566 -> 0.8663Improved
CF vs naive on target accuracy, selection on predictorscounterfactually fair predictor · Selection on Predictorstarget accuracy: 0.8700 -> 0.8699No clear effect
CF vs FTU on target accuracy, measurement errorcounterfactually fair predictor · Measurement Errortarget accuracy: 0.8114 -> 0.8204Improved
CF vs naive on source accuracy, measurement errorcounterfactually fair predictor · Measurement Errorsource accuracy: 0.8283 -> 0.7695Hurt
CF vs naive on source accuracy, selection on labelcounterfactually fair predictor · Selection on Labelsource accuracy: 0.8771 -> 0.8610Hurt
CF target accuracy vs predictor trained directly on target, selection on labelcounterfactually fair predictor · Selection on Labeltarget accuracy: 0.8663 -> 0.8655No clear effect
CF demographic parity difference under measurement errorcounterfactually fair predictor · Measurement Errordemographic parity difference: -0.0005Improved
CF equalized odds difference under selection on labelcounterfactually fair predictor · Selection on Labelequalized odds difference: -0.0021Improved
CF calibration difference under selection on predictorscounterfactually fair predictor · Selection on Predictorscalibration difference: 0.0040Improved
CF calibration difference under measurement error (mismatched metric)counterfactually fair predictor · Measurement Errorcalibration difference: -0.8158Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy