AI Paper Atlas

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

How does a Neural Network's Architecture Impact its Robustness to Noisy Labels?

Jingling Li, Mozhi Zhang, Keyulu Xu, John P Dickerson, Jimmy Ba

NeurIPS 2021 · Label-efficient and robust learning › Noisy labels · approach: Sample selection and co-training · 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
evaluate predictive power of learned representations instead of raw test accuracyDivideMix PreAct ResNet18 · CIFAR-10 flipped noise 80%test accuracy (%): 56.2 -> 93.6Improved
evaluate predictive power of learned representations instead of raw test accuracyDivideMix 4-layer MLP · CIFAR-Easy uniform noise 90%test accuracy (%): 16.22 -> 99.94Improved
use CNN-based network instead of MLP (better alignment with target)DivideMix predictive power · CIFAR-10 uniform noise 80%test accuracy (%): 38.8 -> 93.5Improved
use MLP instead of CNN (better alignment with target)DivideMix predictive power · CIFAR-Easy uniform noise 90%test accuracy (%): 10.09 -> 99.94Improved
add DivideMix on top of vanilla training4-layer MLP · CIFAR-10 uniform noise 80%test accuracy (%): 32.5 -> 34.4No clear effect
add DivideMix on top of vanilla training9-layer CNN · CIFAR-Easy uniform noise 90%test accuracy (%): 10.36 -> 10.00No clear effect
add DivideMix on top of vanilla training9-layer CNN · CIFAR-Easy uniform noise 80%test accuracy (%): 10.15 -> 100.00Improved
add DivideMix on top of vanilla trainingPreAct ResNet18 · CIFAR-10 uniform noise 80%test accuracy (%): 27.3 -> 92.9Improved
add DivideMix on top of vanilla training4-layer MLP · CIFAR-10 flipped noise 80%test accuracy (%): 43.0 -> 44.0No clear effect
use predictive power instead of test accuracy for weakly aligned MLPDivideMix 4-layer MLP · CIFAR-10 flipped noise 80%test accuracy (%): 44.0 -> 38.8Hurt
use predictive power of DivideMix on well-aligned CNN under extreme uniform noise9-layer CNN · CIFAR-Easy uniform noise 90%test accuracy (%): 10.00 -> 10.09No clear effect
increase mutual information between noisy and clean labelswell-aligned networks · predictive power vs noise typetest accuracy (%)Improved

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