AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Learning Antidote Data to Individual Unfairness

Peizhao Li, Ethan Xia, Hongfu Liu

ICML 2023 · Learning theory and algorithms › Fairness · approach: Post-processing and transport for fairness · 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
Add antidote data to NN trainingNN · AdultPos. Comp. Mean: 33.21 -> 26.05Improved
Add antidote data through DRO, selecting worst-loss comparable samplesNN · AdultPos. Comp. Mean: 33.21 -> 17.46Improved
Use antidote data with DRO rather than ordinary antidote-data trainingNN · AdultPos. Comp. Mean: 26.05 -> 17.46Improved
Add antidote data to NN trainingNN · Law SchoolROC: 85.70 -> 86.11Improved
Add antidote data to NN trainingNN · Law SchoolPos. Comp. Mean: 5.38 -> 1.59Improved
Add antidote data to NN trainingNN · DutchPos. Comp. Mean: 15.88 -> 8.71Improved
Add antidote data to NN trainingNN · OuladPos. Comp. Mean: 6.63 -> 4.09Improved
Add antidote data to NN trainingNN · DutchROC: 90.22 -> 87.05Hurt
Add antidote data to NN trainingNN · OuladROC: 65.80 -> 64.75Hurt
Add sensitive-feature removal to antidote-data training (ablation)NN+Anti · AdultPos. Comp. Mean: 26.05 -> 24.40Improved
Use only on-manifold antidote data instead of random comparable samples (ablation)NN training on Adult · AdultPos. Comp. Mean: 27.94 -> 26.05Improved
Train only on antidote data instead of original data (ablation)Adult classifier · AdultAccuracy: 84.64 -> 78.48Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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