AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

ON THE ROLE OF IMPLICIT REGULARIZATION OF STOCHASTIC GRADIENT DESCENT IN GROUP ROBUSTNESS

Nahal Mirzaie, Alireza Alipanah, Ali Abbasi, Amirmahdi Farzane, Hossein Jafarinia, Erfan Sobhaei, Mahdi Ghaznavi, Amir Najafi, Mahdieh Soleymani Baghshah, Mohammad Hossein Rohban

ICLR 2026 · Optimization › Implicit bias and generalization · approach: Implicit regularization of SGD

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
reduce batch size from 256 to 8 (ablation)SGD · CMNIST (rho=5%)worst-group accuracy: 64.7 -> 67.5Improved
reduce batch size from 256 to 8 (ablation)SGD · Domino (rho=5%)worst-group accuracy: 43.4 -> 59.3Improved
reduce batch size from 256 to 8 (ablation)SGD · Waterbirds (rho=5%)worst-group accuracy: 68.2 -> 79.7Improved
reduce batch size from 256 to 8 (ablation)SGD · CelebA (rho=5%)worst-group accuracy: 45.0 -> 46.0Mixed
reduce batch size from 256 to 8 (ablation)SGD · CIFAR10 (rho=5%)worst-group accuracy: 77.9 -> 80.1Improved
reduce batch size from 256 to 8 (ablation)SGD with BERT · Multi-NLIworst-group accuracy: 75.17 -> 76.75Improved
reduce batch size from 256 to 8 (ablation)SGD with BERT · Multi-NLIaccuracy: 79.34 -> 82.18Improved
reduce batch size from 128 to 8 (ablation)SGD with BERT · CivilCommentsworst-group accuracy: 53.70 -> 60.72Improved
increase learning rate while accuracy is near optimal (ablation)SGD · five spurious-correlation benchmarksworst-group accuracy: WGA increases with learning rate, reaches a near-optimal point, then declinesMixed
use small batch size when training explicit debiasing method (ablation)DFR · Domino-CMFworst-group accuracy: 42.7 -> 72.9Improved
use small batch size when training explicit debiasing method (ablation)AFR · Domino-CMFworst-group accuracy: 40.3 -> 67.3Improved
use small batch size when training explicit debiasing method (ablation)EVaLS · Domino-CMFworst-group accuracy: 51.2 -> 76.7Improved

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