AI Paper Atlas

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

Can Less be More? When Increasing-to-Balancing Label Noise Rates Considered Beneficial

Yang Liu, Jialu Wang

NeurIPS 2021 · Label-efficient and robust learning › Noisy labels · approach: Instance-dependent noise modeling · 3 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
Balance class-dependent label noise before training with cross-entropyAdult one-layer perceptron · Adult, e_-=0.0, e_+=0.1accuracy: 72.63±0.29 -> 73.62±0.37Improved
Balance class-dependent label noise before training with peer lossCompas one-layer perceptron · Compas, e_-=0.0, e_+=0.1accuracy: 68.06±0.70 -> 68.22±0.68No clear effect
Balance class-dependent label noise before training with peer lossFairface one-layer perceptron · Fairface, e_-=0.0, e_+=0.1accuracy: 87.58±0.15 -> 88.49±0.12Improved
Balance class noise rates before training with peer lossMNIST MLP · MNIST, noise gap 0.1accuracy: 88.12 -> 89.19Improved
Balance class noise rates before training with cross-entropyMNIST MLP · MNIST, noise gap 0.3accuracy: 85.24 -> 81.89Hurt
Balance class noise rates before training with peer lossCIFAR-10 classifier · CIFAR-10, noise gap 0.1accuracy: 89.66 -> 89.90No clear effect
Balance class noise rates before training with cross-entropyCIFAR-10 classifier · CIFAR-10, noise gap 0.3accuracy: 89.78 -> 87.98Hurt
Balance noise rates before training with cross-entropyClothing1M ResNet-50 · Clothing1Mtest accuracy: 68.94% -> 70.37%Improved
Balance group-dependent noise before fairness-constrained trainingAdult logistic regression · Adult, e_a=0.2, e_b=0.4fairness violation: 6.48 -> 3.16Improved
Balance group-dependent noise before fairness-constrained trainingFairface logistic regression · Fairface, e_a=0.2, e_b=0.4fairness violation: 5.87 -> 1.38Improved

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