AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Trade-offs and Guarantees of Adversarial Representation Learning for Information Obfuscation

Han Zhao, Jianfeng Chi, Yuan Tian, Geoffrey J. Gordon

NeurIPS 2020 · Learning theory and algorithms › Fairness · approach: Minimax fairness via sampling · 6 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
Increase lambda from 0 to 5 (ablation)MAX-ENT · Adult, gendertarget task accuracy: 0.8501 -> 0.7891Hurt
Increase lambda from 0 to 5 (ablation)GRL · Adult, gendertarget task accuracy: 0.8501 -> 0.8387Hurt
Increase lambda from 0 to 5 (ablation)ALT-UP · Adult, gendertarget task accuracy: 0.8501 -> 0.8456Hurt
Increase lambda from 0 to 1 (ablation)ALT-UP · Adult, genderadversary accuracy on gender: 0.7408 -> 0.6627Improved
Increase lambda from 0 to 5 (ablation)MAX-ENT · Adult, genderadversary accuracy on gender: 0.7408 -> 0.5695Improved
Increase lambda from 0 to 1 (ablation)GRL · Adult, ageadversary accuracy on age: 0.7190 -> 0.5361Improved
Increase lambda from 0 to 5 (ablation)MAX-ENT · Adult, agetarget task accuracy: 0.8467 -> 0.7795Hurt
Increase lambda from 0 to 5 (ablation)ALT-UP · Adult, agetarget task accuracy: 0.8467 -> 0.8451No clear effect
Increase lambda from 0 to 5 (ablation)ALT-UP · Adult, educationtarget task accuracy: 0.8494 -> 0.8494No clear effect
Increase lambda from 0 to 5 (ablation)GRL · Adult, educationadversary accuracy on education: 0.7088 -> 0.5528Improved
Increase lambda from 0 to 0.1 (ablation)GRL · Adult, educationtarget task accuracy: 0.8494 -> 0.8525Improved
Increase lambda from 0 to 5 (ablation)MAX-ENT · Adult, educationtarget task accuracy: 0.8494 -> 0.8087Hurt

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