AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Fair Representation: Guaranteeing Approximate Multiple Group Fairness for Unknown Tasks

Xudong Shen, Yongkang Wong, Mohan Kankanhalli

TPAMI 2022 · Learning theory and algorithms › Fairness · approach: Fairness-constrained classification · 20 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
learn fair representation with kernel-based fairness regularizer instead of reconstruction onlyreconstruction autoencoder · Adult (sensitive: gender, target: income)sensitive attribute balanced accuracy: .846 -> .573Improved
same changereconstruction autoencoder · Adult (sensitive: gender, target: income)statistical parity difference: .177 -> .029Improved
same changereconstruction autoencoder · Adult (sensitive: gender, target: income)target attribute balanced accuracy: .811 -> .681Hurt
same changereconstruction autoencoder · Adult (sensitive: gender, target: income)DPC: .103 -> .100No clear effect
use ours instead of FFVAEFFVAE · Adult (sensitive: gender, target: income)target attribute balanced accuracy: .610 -> .733Improved
increase length scale from 2 to 2 sqrt2 (ablation)Ours · Adult (sensitive: gender, target: income)sensitive attribute balanced accuracy: .611 -> .573Improved
increase length scale from 2 to 2 sqrt2 (ablation)Ours · Adult (sensitive: gender, target: income)target attribute balanced accuracy: .733 -> .681Hurt
apply fairness representation learning on MPI3Dreconstruction autoencoder · MPI3D uncorrelated (sensitive: object color)sensitive attribute balanced accuracy: .935 -> .507Improved
apply fairness representation learning on MPI3Dreconstruction autoencoder · MPI3D uncorrelated (sensitive: object color), target shapebalanced accuracy: .804 -> .846Improved
FFVAE on MPI3D background colorreconstruction autoencoder · MPI3D uncorrelated (sensitive: object color), target bkgd colorDOpp: .047 -> .232Hurt
apply gender-blind learning to ArcFaceArcFace · LFW face verificationTAR@1e-3 (%): 94.4 -> 86.8Hurt
apply gender-blind learning to ArcFaceArcFace · VGGFace2 gender prediction (logistic regression)gender prediction (%): 96.9 -> 52.6Improved

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