AI Paper Atlas

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

Robustness to Label Noise Depends on the Shape of the Noise Distribution

Diane Oyen, Michal Kucer, Nick Hengartner, Har Simrat Singh

NeurIPS 2022 · Label-efficient and robust learning › Noisy labels · approach: Instance-dependent noise modeling · 0 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
feature-dependent (GapMax hash-based) instead of class-dependent noise (ablation)ResNet-32 baseline · CIFAR10 eps=0.2, s=1clean accuracy: 78.15 -> 76.92Hurt
feature-dependent instead of class-dependent noise (ablation)ResNet-32 baseline · CIFAR10 eps=0.6, s=5clean accuracy: 48.70 -> 40.70Hurt
feature-dependent instead of class-dependent noise (ablation)ResNet-32 baseline · CIFAR100 eps=0.4, s=1clean accuracy: 42.24 -> 37.17Hurt
add CleanLabResNet-32 · CIFAR10 feature noise eps=0.4, s=1clean accuracy: 59.86 -> 82.47Improved
add MixUpResNet-32 · CIFAR100 feature noise eps=0.4, s=1clean accuracy: 37.17 -> 47.73Improved
add CleanLabResNet-32 · CIFAR100 class noise eps=0.4, s=1clean accuracy: 42.24 -> 32.84Hurt
add Co-TeachingResNet-32 · CIFAR10 class noise eps=0.6, s=1clean accuracy: 35.26 -> 22.80Hurt
add SCEResNet-32 · CIFAR10 class noise eps=0.4, s=1clean accuracy: 58.94 -> 57.42No clear effect
add MixUpResNet-32 · CIFAR10 class noise eps=0.6, s=1clean accuracy: 35.26 -> 12.27Hurt
add CleanLab or CoTeaching at higher noise levels2-hidden-layer NN · synthetic 2D 10-class feature noiseclean accuracyMixed
uniform-x feature noise (constant flip probability) vs class-dependent noise (ablation)2-hidden-layer NN · synthetic 2Dclean accuracyNo clear effect
GapMax worst-case noise vs other noise types (ablation)2-hidden-layer NN · synthetic 2Dclean accuracyHurt

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