AI Paper Atlas

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

Learning With Noisy Labels via Sparse Regularization

Xiong Zhou, Xianming Liu, Chenyang Wang, Deming Zhai, Junjun Jiang, Xiangyang Ji

ICCV 2021 · Label-efficient and robust learning › Noisy labels · approach: Sample selection and co-training · 85 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 sparse regularization (SR) to CECE · MNIST (0.8 symmetric noise)test accuracy: 22.66 -> 98.06Improved
add sparse regularization (SR) to CECE · CIFAR-10 (0.8 symmetric noise)test accuracy: 19.55 -> 51.13Improved
add sparse regularization (SR) to CECE · CIFAR-100 (0.8 symmetric noise)test accuracy: 7.59 -> 17.35Improved
add sparse regularization (SR) to CECE · CIFAR-100 (clean)test accuracy: 71.33 -> 72.19Improved
add SR to FLFL · WebVision (Mini, ResNet-50, top-1 val)top-1 accuracy: 63.80 -> 70.28Improved
add SR to CECE · WebVision (Mini, ResNet-50, top-1 val)top-1 accuracy: 66.96 -> 69.12Improved
add SR to CECE · CIFAR-10 (0.4 asymmetric noise)test accuracy: 74.67 -> 79.29Improved
add SR to CE for imbalanced classificationCE · CIFAR-10 (long-tailed 0.01)test accuracy: 64.16 -> 69.78Improved
add SR to CE for imbalanced classificationCE · CIFAR-100 (long-tailed 0.01)test accuracy: 35.17 -> 41.24Improved
use GCE or APL instead of CE+SR under heavy noiseCE+SR · CIFAR-100 (0.8 symmetric noise)test accuracy: 17.35 -> 24.91Hurt
use GCE instead of CE on clean dataCE · CIFAR-100 (clean)test accuracy: 71.33 -> 63.09Hurt
increase lambda of l_p regularization to 7 or more (with tau=1, p=0.1) (ablation)CE+SR · CIFAR-10 (0.6 symmetric noise)test accuracyMixed

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