AI Paper Atlas

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

Dimensionality-Driven Learning with Noisy Labels

Xingjun Ma, Yisen Wang, Michael E. Houle, Shuo Zhou, Sarah Erfani, Shutao Xia, Sudanthi Wijewickrema, James Bailey

ICML 2018 · Label-efficient and robust learning › Noisy labels · approach: Sample selection and co-training · 240 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
Use D2L's LID-triggered adaptive interpolation of raw and predicted labels instead of the strongest reported baseline, ForwardLeNet-5 · MNIST, 20%–60% symmetric label noisetest accuracy (%): 88.02±0.1 / 68.46±0.1 / 45.51±0.2 -> 98.84±0.1 / 98.49±0.1 / 94.73±0.2 (cross-entropy); Forward: 96.45±0.1 / 94.90±0.1 /Improved
Use D2L's LID-triggered adaptive interpolation of raw and predicted labels instead of Forward6-layer CNN · SVHN, 20%–60% symmetric label noisetest accuracy (%): 87.63±0.1 / 82.68±0.1 / 80.92±0.2 vs 85.51±0.1 / 79.09±0.2 / 62.57±0.2Improved
Use D2L's LID-triggered adaptive interpolation of raw and predicted labels instead of Forward12-layer CNN · CIFAR-10, 20%–60% symmetric label noisetest accuracy (%): 85.13±0.2 / 83.36±0.3 / 72.84±0.3 vs 84.61±0.3 / 82.84±0.2 / 72.41±0.4Improved
Use D2L's LID-triggered adaptive interpolation of raw and predicted labels instead of ForwardResNet-44 · CIFAR-100, 20%–60% symmetric label noisetest accuracy (%): 62.20±0.4 / 52.01±0.3 / 42.27±0.2 vs 60.25±0.2 / 51.27±0.3 / 41.22±0.3Improved
Use D2L rather than ForwardLeNet-5 · MNIST, clean labelstest accuracy (%): 99.30±0.0 -> 99.28±0.0Hurt
Use D2L rather than Forward12-layer CNN · CIFAR-10, clean labelstest accuracy (%): 90.27±0.1 -> 89.41±0.2Hurt
Monitor LID during training and adjust the loss after dimensionality expansion beginsD2L on CIFAR-10 · CIFAR-10 with 40% and 60% noisy labelstest accuracyImproved
Use LID-based loss adaptation to regulate trainingD2L on CIFAR-10 · CIFAR-10 with 40% and 60% noisy labelsrepresentation subspace dimensionality (LID)Improved
Vary neighborhood size k and batch count m for LID estimation (ablation)D2L on CIFAR-10 · CIFAR-10, varying noise ratestest accuracyNo clear effect

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