AI Paper Atlas

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

Learning Fast Sample Re-Weighting Without Reward Data

Zizhao Zhang, Tomas Pfister

ICCV 2021 · Label-efficient and robust learning › Noisy labels · approach: Noise-aware loss and pairwise methods · 106 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 partial layers (feature sharing) for the meta model (ablation)FSR · CIFAR10 40% uniform noise, ResNet32accuracyImproved
use partial meta layers (feature sharing) instead of full modelFSR vs L2R · training costtraining time overhead: 47% ~ 87% vs 319% ~ 443%Improved
use partial meta layers instead of full modelFSR vs L2R · GPU memorymemory overhead: 1.9% ~ 34% vs 54% ~ 270%Improved
add MixUp for the weighted loss (ablation)FSR · noise robust trainingdictionary purity and accuracyImproved
remove MixUp from forgetting-event pusher (ablation)FSR (forget-event pusher) · CIFAR100 20%/40% noiseaccuracy: 68.1 / 60.9Improved
add momentum re-labeling (ablation)FSR · CIFAR10 20% uniform noiseaccuracy, purityImproved
use meta-margin instead of negative-loss or AUM pusher (ablation)FSR · CIFAR100 20%/40% noiseaccuracyImproved
use forgetting-event inverse as pusher instead of meta-margin (ablation)FSR · CIFAR100 noiseaccuracyHurt
add MixUp or label smoothing for long-tailed recognition (ablation)FSR · CIFAR long-tailedaccuracyNo clear effect
increase real reward data size (ablation)L2R · CIFAR10 40% uniform noiseaccuracyNo clear effect
vary dictionary size (ablation)FSR · CIFAR10 40% uniform noiseaccuracyNo clear effect
sweep hyper-parameters (ablation)FSR · CIFAR100 40% uniform noiseaccuracy: ~2%No 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