AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Label-Imbalanced and Group-Sensitive Classification under Overparameterization

Ganesh Ramachandra Kini, Orestis Paraskevas, Samet Oymak, Christos Thrampoulidis

NeurIPS 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Ranking-based loss optimization · 10 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 multiplicative logit adjustment to LA-loss (VS vs LA) (ablation)LA-loss · CIFAR-10 STEP-100top-1 accuracy (%): 78.23 -> 79.10Improved
add additive logit adjustment to CDT-loss (VS vs CDT) (ablation)CDT-loss · CIFAR-10 STEP-100top-1 accuracy (%): 73.26 -> 79.10Improved
add multiplicative adjustment to LA-loss (VS vs LA) (ablation)LA-loss · CIFAR-10 LT-100top-1 accuracy (%): 80.81 -> 80.82No clear effect
add multiplicative adjustment to LA-loss (VS vs LA) (ablation)LA-loss · CIFAR-100 LT-100top-1 accuracy (%): 42.87 -> 43.52Improved
add multiplicative adjustment to LA-loss (VS vs LA) (ablation)LA-loss · CIFAR-100 STEP-100top-1 accuracy (%): 45.69 -> 46.53Improved
group-sensitive VS-loss instead of CEResNet50 · Waterbirdsworst-group accuracy (%): 68.1 -> 76.7Improved
add DRO to Group VS (ablation)Group VS · Waterbirdsworst-group accuracy (%): 76.7 -> 78.9Improved
use additive-only adjustments (Group LA) instead of CEResNet50 · Waterbirdsbalanced accuracy (%): 84.9 -> 84.2No clear effect
add multiplicative adjustments (Group CDT instead of Group LA) (ablation)Group LA · Waterbirdsworst-group accuracy (%): 70.1 -> 75.4Improved
add additive adjustments to Group CDT (Group VS vs Group CDT) (ablation)Group CDT · Waterbirdsworst-group accuracy (%): 75.4 -> 76.7No clear effect
add additive adjustments to CDT with tau=-0.5 and train 300 epochs (ablation)CDT-loss (gamma=0.15) · CIFAR-10 LT-100train accuracy at 200 epochs (%): ~93 -> ~97Improved
add additive adjustments to CDT when training to interpolation (300 epochs) (ablation)CDT-loss · CIFAR-10 LT-100balanced test accuracyNo 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