AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Identifying Hard Noise in Long-Tailed Sample Distribution

Xuanyu Yi, Kaihua Tang, Xian-Sheng Hua, Joo-Hwee Lim, Hanwang Zhang

ECCV 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Classifier rebalancing and margin losses · 22 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 environment-invariant hard noise identification (H2E) vs MentorMix-RSH2E · ImageNet-NLT red 10% noiseTop-1 accuracy (%): 62.20 -> 64.86Improved
use H2E instead of MentorMix-RSH2E · ImageNet-NLT blue 30% noiseTop-1 accuracy (%): 48.27 -> 51.52Improved
iterate hard-to-easy transformation (H2E-iter)H2E · ImageNet-NLT red 30% noiseTop-1 accuracy (%): 58.38 -> 60.66Improved
add augmentation to construct environments (3 environments) (ablation)H2E · ImageNet-NLT red 10% noiseTop-1 accuracy (%): 62.79 -> 64.86Improved
increase number of environments from 2 to 3 (ablation)H2E · ImageNet-NLT red 10% noise (with augmentation)Top-1 accuracy (%): 62.15 -> 64.86Improved
increase number of environments from 3 to 4 (ablation)H2E · ImageNet-NLT red 10% noise (with augmentation)Top-1 accuracy (%): 64.86 -> 65.38No clear effect
use H2E as stage 2 instead of CE+RW after CF stage 1 (ablation)H2E · ImageNet-NLT red 10% noiseTop-1 accuracy (%): 56.54 -> 63.90Improved
use H2E stage 1 instead of LDAM before H2E stage 2 (ablation)H2E · ImageNet-NLT red 10% noiseTop-1 accuracy (%): 61.94 -> 64.86Improved
use H2E stage 2 instead of ERM+RW after H2E stage 1 (ablation)H2E · ImageNet-NLT red 10% noiseTop-1 accuracy (%): 60.20 -> 64.86Improved
use H2E instead of LDAM on Animal10-NLTH2E · Animal10-NLT (eta=100)Top-1 accuracy (%): 68.21 -> 66.58No clear effect
use H2E instead of BBN on Food101-NLTH2E · Food101-NLT (eta=100)Top-1 accuracy (%): 53.16 -> 58.66Improved
use red (realistic) instead of blue (synthetic) noiseCE · ImageNet-NLT 10% noiseTop-1 accuracy (%): 45.80 -> 54.36Improved

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