AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

Mitigating Memorization of Noisy Labels by Clipping the Model Prediction

Hongxin Wei, HUIPING ZHUANG, RENCHUNZI XIE, Lei Feng, Gang Niu, Bo An, Yixuan Li

ICML 2023 · Label-efficient and robust learning › Noisy labels · approach: Sample selection and co-training · 1 citations (OpenAlex)

Read the paper

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 LogitClip to CE lossCE · CIFAR-10 instance-dependent noisetest accuracy: 68.26 to 86.74Improved
add LogitClip to CE lossCE · CIFAR-10 symmetric-50% and other noise typestest accuracy: e.g. 70.88 to 84.37 (Sym-50%)Improved
add LogitClip to CE lossCE · CIFAR-100 noise settingstest accuracy: e.g. 47.07 to 63.16 (Sym-50%)Improved
add LogitClip to robust losses (MAE, PHuber-CE, SCE, GCE, Taylor-CE, NCE, AEL, AUL, Cores, NCE+MAE, NCE+AGCE)robust losses · CIFAR-10 and CIFAR-100test accuracy: large gains on many settings (e.g. AUL Sym-20% 76.73 to 91.31) but near-zero or slight drops on some (SCE Sym-20% 91.48 Mixed
add LogitClip to CE lossCE on WebVision Mini ResNet-18 · WebVision Minitop-1 validation accuracy (best epoch): 62.6 to 65.12Improved
add LogitClip to DivideMixDivideMix · CIFAR-10 noisy settingstest accuracy: 94.41 to 95.15 (Sym-50%)Improved
clip logit norm with tau small (0.25–1) instead of large tau on clean CIFAR-100 (ablation)CE + LogitClip · clean CIFAR-100accuracy: 75.98 (CE) to 46.04 at tau=0.25, 68.00 at tau=0.5Hurt
use a large tau (20) for LogitClip on clean CIFAR-100 (ablation)CE + LogitClip · clean CIFAR-100accuracy: 75.90 vs CE 75.98No clear effect
replace norm clipping with clipping-by-value on logits (ablation)CE · noisy CIFAR-10 (instance-dependent, real)test accuracyHurt
replace LogitClip with LogitNorm (exact norm constraint) (ablation)CE · CIFAR-10 asymmetric noisetest accuracy: 81.81 (LogitNorm) vs 86.91 (LogitClip)Hurt
add Norm Regularization (L2 logit penalty) instead of LogitClip (ablation)CE · noisy CIFAR-10 across noise typestest accuracyHurt
apply ReLU6-style clamping of intermediate activations (ablation)CE · noisy CIFAR-10test accuracyNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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