AI Paper Atlas

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

Learning from Multiple Annotator Noisy Labels via Sample-Wise Label Fusion

Zhengqi Gao, Fan-Keng Sun, Mingran Yang, Sucheng Ren, Zikai Xiong, Marc Engeler, Antonio Burazer, Linda Wildling, Luca Daniel, Duane S. Boning

ECCV 2022 · Label-efficient and robust learning › Noisy labels · approach: Noise-aware loss and pairwise methods · 10 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 sample-wise confusion matrices to the weight-vector-only model (ablation)Ours · CIFAR-100, eps=22accuracy: 63.59 -> 64.48Improved
add sample-wise confusion matrices to the weight-vector-only model (ablation)Ours · CIFAR-100, eps=23accuracy: 59.40 -> 60.12Improved
add sample-wise weight vectors to the confusion-matrix-only model (ablation)Ours · CIFAR-100, eps=22accuracy: 60.25 -> 64.48Improved
add sample-wise weight vectors to the confusion-matrix-only model (ablation)Ours · CIFAR-100, eps=23accuracy: 57.07 -> 60.12Improved
replace majority voting with sample-wise label fusionOurs vs Mjv · MNIST, eps=30accuracy: 87.72 -> 92.49Improved
replace the second best method with sample-wise label fusionOurs · MNIST, eps=30accuracy: 4.77% more accuracyImproved
replace the second best method with sample-wise label fusionOurs · CIFAR-100, eps=22accuracy: 5.74% more accuracyImproved
replace WDN with sample-wise label fusionOurs · ImageNet-100, eps=580accuracy: 56.03 -> 73.34Improved
replace MBEM with sample-wise label fusionOurs · ImageNet-100, eps=680accuracy: 20.06 -> 54.93Improved
replace the best baseline with sample-wise label fusionOurs · CIFAR-100 with pretrained neural annotatorsaccuracy: 72.08 -> 73.14Improved
compare sample-wise label fusion with MBEM when 3 annotators are always correctOurs · MNIST, N_correct=3accuracy: 88.31% vs 88.26%No clear effect
vary lambda of the diagonal regularizer (ablation)Ours · CIFAR-100, eps=22/23accuracy: best at lambda=1.0Mixed

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