AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Regularized Discriminative Alignment for Deep Representations under Label Shift

Hengchao Shi, Boen Jiang, Guanhua Fang, Wen Yu, Ming Zheng

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory

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
use LDA instead of PCA for summary functions (ablation)RDALS · MNISTweight estimation MSE: 2.938 -> 0.018Improved
use LDA instead of random projection for summary functions (ablation)RDALS · MNISTweight estimation MSE: 2.195 -> 0.018Improved
use LDA instead of PCA for summary functions (ablation)RDALS · CIFAR-10weight estimation MSE: 0.573 -> 0.051Improved
use LDA instead of random projection for summary functions (ablation)RDALS · CIFAR-10weight estimation MSE: 0.736 -> 0.051Improved
use LDA instead of PCA for summary functions (ablation)RDALS · CIFAR-100weight estimation MSE: 3.248 -> 1.686Improved
use LDA instead of random projection for summary functions (ablation)RDALS · CIFAR-100weight estimation MSE: 4.063 -> 1.686Improved
use regularized SOCP solver instead of direct inverse (LDA) (ablation)RDALS · CIFAR-10weight estimation MSE: 4.785 -> 0.056Improved
use regularized SOCP solver instead of direct inverse (LDA) (ablation)RDALS · MNISTweight estimation MSE: 3.879 -> 0.014Improved
use regularized SOCP solver instead of direct inverse (LDA) (ablation)RDALS · CIFAR-100weight estimation MSE: 29.325 -> 1.703Improved
use regularized SOCP solver instead of direct inverse (PCA) (ablation)RDALS with PCA · CIFAR-10weight estimation MSE: 7.664 -> 0.489Improved
vary regularization lambda from 1e-6 to 1e-1 (ablation)RDALS · CIFAR-10 (N=10000)weight estimation MSE: 4.555e-2 -> 4.857e-2No clear effect
RDALS versus CPMCN running timeRDALS vs CPMCN · CIFAR-10 (N=10000)running time (s): 167.82 -> 74.18Improved

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