AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Adaptive Data Debiasing through Bounded Exploration

Yifan Yang, Yang Liu, Parinaz Naghizadeh

NeurIPS 2022 · 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 bounded exploration instead of exploitation-only data collectionActive debiasing · Classification with censored feedbackdistribution-estimation biasImproved
Use exploitation-only data collection with no explorationDistribution estimation · Classification with censored feedbackdistribution-estimation biasHurt
Explore below the decision threshold without a lower boundPure exploration · Classification with censored feedbackdistribution-estimation biasImproved
Replace unbounded pure exploration with bounded explorationActive debiasing · Gaussian synthetic dataspeed of debiasingHurt
Explore more deeply below the classifierActive debiasing and pure exploration · Gaussian synthetic dataregretImproved
Explore more deeply below the classifierActive debiasing and pure exploration · Gaussian synthetic dataweighted regretHurt
Use a higher exploration percentile, limiting the exploration depth (ablation)Active debiasing · Gaussian synthetic dataweighted regretImproved
Apply a fairness constraint that over-selects a groupActive debiasing · Groups subject to fairness constraintsspeed of debiasingHurt
Apply a fairness constraint that under-selects a groupActive debiasing · Groups subject to fairness constraintsspeed of debiasingImproved
Use dimension reduction to map Adult features to a one-dimensional score (ablation)Logistic regression · Adult, overallaccuracy: 82.96% -> 78.07%Hurt
Use dimension reduction to map Adult features to a one-dimensional score (ablation)Logistic regression · Adult, advantaged groupaccuracy: 82.13% -> 76.95%Hurt
Use dimension reduction to map Adult features to a one-dimensional score (ablation)Logistic regression · Adult, disadvantaged groupaccuracy: 88.26% -> 84.69%Hurt

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