TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Image Classification	mini WebVision 1.0	NCR+Mixup+DA (ResNet-50)	Top-1 Accuracy	80.5	# 7
Image Classification	mini WebVision 1.0	NCR+Mixup (ResNet-50)	Top-1 Accuracy	79.4	# 14
Image Classification	mini WebVision 1.0	NCR (ResNet-50)	Top-1 Accuracy	77.1	# 30
Learning with noisy labels	Red MiniImageNet 20% label noise	NCR (ResNet-18)	Test Accuracy	69.0	# 1
Image Classification	Red MiniImageNet 20% label noise	NCR (ResNet-18)	Accuracy	69.0	# 1
Learning with noisy labels	Red MiniImageNet 40% label noise	NCR (ResNet-18)	Test Accuracy	64.6	# 1
Image Classification	Red MiniImageNet 40% label noise	NCR (ResNet-18)	Accuracy	64.6	# 1
Image Classification	Red MiniImageNet 80% label noise	NCR (ResNet-18)	Accuracy	51.2	# 1
Learning with noisy labels	Red MiniImageNet 80% label noise	NCR (ResNet-18)	Test Accuracy	51.2	# 1
Image Classification	WebVision-1000	NCR (ResNet-50)	Top-1 Accuracy	75.7%	# 8
Image Classification	WebVision-1000	NCR+Mixup+DA (ResNet-50)	Top-1 Accuracy	76.8	# 4

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/learning-with-noisy-labels-on-red)](https://paperswithcode.com/sota/learning-with-noisy-labels-on-red?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/image-classification-on-red-miniimagenet-20)](https://paperswithcode.com/sota/image-classification-on-red-miniimagenet-20?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/learning-with-noisy-labels-on-red-1)](https://paperswithcode.com/sota/learning-with-noisy-labels-on-red-1?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/image-classification-on-red-miniimagenet-40)](https://paperswithcode.com/sota/image-classification-on-red-miniimagenet-40?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/image-classification-on-red-miniimagenet-80)](https://paperswithcode.com/sota/image-classification-on-red-miniimagenet-80?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/learning-with-noisy-labels-on-red-3)](https://paperswithcode.com/sota/learning-with-noisy-labels-on-red-3?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/image-classification-on-webvision-1000)](https://paperswithcode.com/sota/image-classification-on-webvision-1000?p=learning-with-neighbor-consistency-for-noisy-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/learning-with-neighbor-consistency-for-noisy-1/image-classification-on-mini-webvision-1-0)](https://paperswithcode.com/sota/image-classification-on-mini-webvision-1-0?p=learning-with-neighbor-consistency-for-noisy-1)`

Learning with Neighbor Consistency for Noisy Labels

CVPR 2022 · Ahmet Iscen, Jack Valmadre, Anurag Arnab, Cordelia Schmid ·

Recent advances in deep learning have relied on large, labelled datasets to train high-capacity models. However, collecting large datasets in a time- and cost-efficient manner often results in label noise. We present a method for learning from noisy labels that leverages similarities between training examples in feature space, encouraging the prediction of each example to be similar to its nearest neighbours. Compared to training algorithms that use multiple models or distinct stages, our approach takes the form of a simple, additional regularization term. It can be interpreted as an inductive version of the classical, transductive label propagation algorithm. We thoroughly evaluate our method on datasets evaluating both synthetic (CIFAR-10, CIFAR-100) and realistic (mini-WebVision, WebVision, Clothing1M, mini-ImageNet-Red) noise, and achieve competitive or state-of-the-art accuracies across all of them.

PDF Abstract CVPR 2022 PDF CVPR 2022 Abstract

Code

Add Remove Mark official

google-research/scenic official

↳ Quickstart in

Colab

3,248

Tasks

Add Remove

Image Classification

Learning with noisy labels

Datasets

CIFAR-10

CIFAR-100 mini-Imagenet

Clothing1M

WebVision Red MiniImageNet 20% label noise Red MiniImageNet 40% label noise Red MiniImageNet 80% label noise

Results from the Paper

Add Remove

Ranked #1 on Image Classification on Red MiniImageNet 80% label noise

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Image Classification	mini WebVision 1.0	NCR+Mixup+DA (ResNet-50)	Top-1 Accuracy	80.5	# 7	Compare
Image Classification	mini WebVision 1.0	NCR+Mixup (ResNet-50)	Top-1 Accuracy	79.4	# 14	Compare
Image Classification	mini WebVision 1.0	NCR (ResNet-50)	Top-1 Accuracy	77.1	# 30	Compare
Learning with noisy labels	Red MiniImageNet 20% label noise	NCR (ResNet-18)	Test Accuracy	69.0	# 1	Compare
Image Classification	Red MiniImageNet 20% label noise	NCR (ResNet-18)	Accuracy	69.0	# 1	Compare
Learning with noisy labels	Red MiniImageNet 40% label noise	NCR (ResNet-18)	Test Accuracy	64.6	# 1	Compare
Image Classification	Red MiniImageNet 40% label noise	NCR (ResNet-18)	Accuracy	64.6	# 1	Compare
Image Classification	Red MiniImageNet 80% label noise	NCR (ResNet-18)	Accuracy	51.2	# 1	Compare
Learning with noisy labels	Red MiniImageNet 80% label noise	NCR (ResNet-18)	Test Accuracy	51.2	# 1	Compare
Image Classification	WebVision-1000	NCR (ResNet-50)	Top-1 Accuracy	75.7%	# 8	Compare
Image Classification	WebVision-1000	NCR+Mixup+DA (ResNet-50)	Top-1 Accuracy	76.8	# 4	Compare

Methods

Add Remove

No methods listed for this paper. Add relevant methods here

Edit Social Preview

Learning with Neighbor Consistency for Noisy Labels

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit Add Remove

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Add Remove

Methods

Add Remove