AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

WCLD: Curated Large Dataset of Criminal Cases from Wisconsin Circuit Courts

Elliott Ash, Naman Goel, Nianyun Li, Claudia Marangon, Peiyao Sun

NeurIPS 2023 · Learning theory and algorithms › Fairness · approach: Fairness auditing and regularization · 0 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
use XGBoost instead of logistic regressionrecidivism classifier · WCLD 180-day cut-off, overallaccuracy: 0.6461 -> 0.6588No clear effect
use XGBoost instead of logistic regressionrecidivism classifier · WCLD 180-day cut-off, overallAUC: 0.6822 -> 0.7039Improved
use XGBoost instead of logistic regressionrecidivism classifier · WCLD 180-day cut-off, overallFPR: 0.1532 -> 0.2244Hurt
use XGBoost instead of logistic regressionrecidivism classifier · WCLD 180-day cut-off, overallFNR: 0.6282 -> 0.5008Improved
use XGBoost instead of logistic regressionrecidivism classifier · WCLD 180-day cut-off, Native American vs HispanicPR gap between groups: 0.4800 vs 0.2910 (XGBoost); 0.3761 vs 0.1991 (LR)Hurt
evaluate Native American group vs Hispanic groupXGBoost · WCLD 180-day cut-offFPR: 0.3272 vs 0.2016Hurt
evaluate African American group vs Caucasian groupXGBoost · WCLD 180-day cut-offFPR: 0.2159 -> 0.2454Hurt
evaluate African American group vs Caucasian groupXGBoost · WCLD 180-day cut-offFNR: 0.5113 -> 0.4792Improved
change recidivism label from 180-day to 2-year sentence cut-offXGBoost · WCLD overallAUC: 0.7039 -> 0.7018No clear effect
change recidivism label from 180-day to 2-year sentence cut-offlogistic regression · WCLD overallFNR: 0.6282 -> 0.6429No clear effect
change recidivism label from 180-day to 2-year sentence cut-offXGBoost · WCLD Native American groupFPR: 0.3272 -> 0.3223No clear effect

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