AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Revisiting (Un)Fairness in Recourse by Minimizing Worst-Case Social Burden

Ainhize Barrainkua, Giovanni De Toni, Jose A. Lozano, Novi Quadrianto

AAAI 2026 · Learning theory and algorithms › Fairness · approach: Causal and policy fairness in decisions · 1 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
train classifier with MISOB burden reweightingfeed-forward NN + GS recourse · ADULT (race groups)worst-group social burden: 4.56 -> 3.01Improved
train classifier with MISOB burden reweightingfeed-forward NN + WT recourse · ADULT (race groups)worst-group social burden: 1.28 -> 0.79Improved
train classifier with MISOB burden reweightingfeed-forward NN + CCHVAE recourse · ADULT (gender groups)worst-group social burden: 6.43 -> 4.18Improved
train classifier with MISOB burden reweightingfeed-forward NN · ADULToverall accuracy: 0.81 -> 0.82No clear effect
train classifier with MISOB burden reweightingfeed-forward NN + GS recourse · ADULT (race groups)worst-group TPR: 0.27 -> 0.52Improved
train classifier with MISOB burden reweightingfeed-forward NN + GS recourse · ADULT (race groups)gap in social burden: 0.03 -> 0.85Hurt
enforce TPR parity with POSTPRO post-processingfeed-forward NN + GS recourse · ADULT (gender groups)worst-group social burden: 4.73 -> 24.46Hurt
enforce TPR parity with POSTPRO post-processingfeed-forward NN + CCHVAE recourse · ADULT (race groups)worst-group social burden: 6.25 -> 11.20Hurt
increase alpha in burden weighting (ablation)MISOB · ADULT (WT/GS, race/gender)worst-group metricsImproved
increase alpha above 0.5 (ablation)MISOB · ADULT (WT/GS, race/gender)overall accuracyHurt

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