AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Interpretable algorithmic fairness in structured and unstructured data

Hari Bandi, Dimitris Bertsimas, Thodoris Koukouvinos, Sofie Kupiec

JMLR 2024 · 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
add meritocracy constraint with delta = 0.1 (ablation)label-flipping method · LSACMerit(lsat): 0.181 -> 0.089Improved
add meritocracy constraint with delta = 0.1 (ablation)label-flipping method · LSACSPD: 0.011 -> 0.072Hurt
add meritocracy constraint with delta = 0.1 (ablation)label-flipping method · CrimeMerit(pct-under-pov): 0.037 -> 0.023Improved
add meritocracy constraint with delta = 0.1 (ablation)label-flipping method · COMPASMerit(priors): 0.076 -> 0.034Improved
add meritocracy constraint with delta = 0.1 (ablation)label-flipping method · German creditMerit(credit): 0.019 -> 0.015Improved
relax meritocracy tolerance from delta = 0.1 to delta = 0.5 (ablation)label-flipping method · LSACSPD: 0.072 -> 0.026Improved
apply label-flipping method (no merit) over nominal modelResNet-18 / tabular classifiers · LSACSPD: 0.299 -> 0.011Improved
apply label-flipping method (no merit) over nominal modeltabular classifier · Crimeaccuracy: 0.779 -> 0.678Hurt
apply label-flipping method over nominal modelResNet-18 · CelebA Brown HairSPD: 0.035 -> 0.011Improved
apply label-flipping method over nominal modelResNet-18 · CelebA Blond HairSPD: 0.059 -> 0.041Improved
alleviate bias jointly for race and gender instead of separately (ablation)label-flipping method · LSACSPD(WM,BM): 0.198 -> 0.007Improved
alleviate bias jointly for race and gender instead of separately (ablation)label-flipping method · LFW Brown HairSPD(WM,BM): 0.174 -> 0.127Improved

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