AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

The Measure and Mismeasure of Fairness

Sam Corbett-Davies, Johann D. Gaebler, Hamed Nilforoshan, Ravi Shroff, Sharad Goel

JMLR 2023 · Learning theory and algorithms › Fairness · approach: Post-processing and transport for fairness

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
enforce demographic parity with group-specific thresholds instead of one thresholddiabetes screening policy · Diabetes screening (NHANES)screening threshold: 1.5% -> 2.2% for Asian Americans and 1% for White AmericansHurt
use a single 1.5% threshold, which violates equal false positive ratesdiabetes screening policy · Diabetes screening (NHANES)false positive rate: 79% for Asian Americans and 67% for White AmericansMixed
remove race from the diabetes risk model (race-blind)risk model · Diabetes screening (NHANES)effective screening threshold: nominal 1.5% acts as 1% for White patients and 3% for Asian patientsHurt
remove gender from the recidivism risk score (gender-blind)COMPAS · Recidivism, Broward Countyrecidivism rate at score 7: 55% for women vs 65% for menHurt

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