AI Paper Atlas

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

Deep Self-Learning From Noisy Labels

Jiangfan Han, Ping Luo, Xiaogang Wang

ICCV 2019 · Label-efficient and robust learning › Noisy labels · approach: Sample selection and co-training · 340 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 self-learning with multi-prototype label correction to cross entropy trainingResNet50 · Clothing1M (1M noisy)accuracy (%): 69.54 -> 74.45Improved
add self-learning label correctionResNet50 · Food101-Naccuracy (%): 84.51 -> 85.11Improved
add self-learning label correction with 25k verification labels for prototype selectionResNet50 · Clothing1M (1M noisy + 25k verify)accuracy (%): 74.69 -> 76.44Improved
add label correction when 50k clean labels are also usedResNet50 · Clothing1M (1M noisy + 50k clean)accuracy (%): 80.27 -> 81.16Improved
use multiple class prototypes (two instead of one) (ablation)SMP · Clothing1Maccuracy (%): +2.04%Improved
combine original noisy label with corrected label (alpha=0.5) instead of noisy only or corrected only (ablation)SMP · Clothing1Maccuracy (%)Improved
use only corrected labels (alpha=1) instead of joint (ablation)SMP · Clothing1Maccuracy (%)Hurt
increase prototype-selection sample count m from 320 to 2560 (ablation)SMP · Clothing1M (1M noisy)accuracy (%): 74.37 -> 74.27No clear effect
use proposed density/cosine prototype selection instead of K-means++ (ablation)SMP · Clothing1M (1M noisy)accuracy (%): 74.08 -> 74.45Improved
use cosine similarity instead of Euclidean distance density peak (ablation)SMP · Clothing1M (1M noisy + 25k verify)accuracy (%): 76.05 -> 76.44Improved
run label correction iteratively through training (ablation)SMP · Clothing1M label accuracylabel accuracy (%): 74.38 -> 77.36Improved

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