AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Benchmarking Stochastic Approximation Algorithms for Fairness-Constrained Training of Deep Neural Networks

Andrii Kliachkin, Jana Lepšová, Gilles Bareilles, Jakub Marecek

ICLR 2026 · Learning theory and algorithms › Fairness · approach: Fairness-constrained classification · 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
train with fairness constraints (SSL-ALM) instead of unconstrained SGDMLP on ACSIncome · Oklahoma binary race, testindependence, separation, sufficiency: independence 0.09837 -> 0.08256; separation 0.15485 -> 0.10806; sufficiency 0.06096 -> 0.04598Improved
train with fairness constraints (SSL-ALM) instead of unconstrained SGDMLP on ACSIncome · Oklahoma binary race, testinaccuracy: 0.20878 -> 0.22321Hurt
use Stochastic Ghost constrained trainingMLP on ACSIncome · Oklahoma binary race, testinaccuracy: 0.20878 -> 0.23896Hurt
use Switching Subgradient constrained trainingMLP on ACSIncome · Oklahoma binary race, testfairness metrics: independence 0.09837 -> 0.10275No clear effect
add Fairret fairness regularizer to SGDMLP on ACSIncome · Oklahoma binary race, testindependence, separation: independence 0.09837 -> 0.09064No clear effect
raise Fairret penalty multiplier from 0 to 3 (ablation)SGD-Fairret · Oklahoma validationloss vs constraint violation: loss 0.406 -> 0.539; violation 0.020 -> 0.004Mixed
use constrained algorithms instead of penalized SGDMLP on ACSIncome · Virginia multi-valued attributeloss and constraint satisfactionNo clear effect
use constrained algorithms instead of penalized SGDMLP on ACSIncome · Virginia multi-valued attributehyperparameter sensitivityImproved
remove smoothing term (ALM vs SSL-ALM) (ablation)SSL-ALM · Oklahoma validationloss: 0.4465 -> 0.4448No clear effect

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