TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Long-tail Learning	CIFAR-100-LT (ρ=10)	MiSLAS	Error Rate	36.8	# 18
Long-tail Learning	CIFAR-100-LT (ρ=100)	MiSLAS	Error Rate	53	# 37
Long-tail Learning	CIFAR-100-LT (ρ=50)	MiSLAS	Error Rate	47.7	# 19
Long-tail Learning	CIFAR-10-LT (ρ=10)	MiSLAS	Error Rate	10	# 18
Long-tail Learning	CIFAR-10-LT (ρ=100)	MiSLAS	Error Rate	17.9	# 16
Long-tail Learning	ImageNet-LT	MiSLAS	Top-1 Accuracy	52.7	# 44
Long-tail Learning	iNaturalist 2018	MiSLAS	Top-1 Accuracy	71.6%	# 25

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-cifar-10-lt-r-100)](https://paperswithcode.com/sota/long-tail-learning-on-cifar-10-lt-r-100?p=improving-calibration-for-long-tailed-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-cifar-100-lt-r-10)](https://paperswithcode.com/sota/long-tail-learning-on-cifar-100-lt-r-10?p=improving-calibration-for-long-tailed-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-cifar-10-lt-r-10)](https://paperswithcode.com/sota/long-tail-learning-on-cifar-10-lt-r-10?p=improving-calibration-for-long-tailed-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-cifar-100-lt-r-50)](https://paperswithcode.com/sota/long-tail-learning-on-cifar-100-lt-r-50?p=improving-calibration-for-long-tailed-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-inaturalist-2018)](https://paperswithcode.com/sota/long-tail-learning-on-inaturalist-2018?p=improving-calibration-for-long-tailed-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-cifar-100-lt-r-100)](https://paperswithcode.com/sota/long-tail-learning-on-cifar-100-lt-r-100?p=improving-calibration-for-long-tailed-1)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/improving-calibration-for-long-tailed-1/long-tail-learning-on-imagenet-lt)](https://paperswithcode.com/sota/long-tail-learning-on-imagenet-lt?p=improving-calibration-for-long-tailed-1)`

Improving Calibration for Long-Tailed Recognition

CVPR 2021 · Zhisheng Zhong, Jiequan Cui, Shu Liu, Jiaya Jia ·

Deep neural networks may perform poorly when training datasets are heavily class-imbalanced. Recently, two-stage methods decouple representation learning and classifier learning to improve performance. But there is still the vital issue of miscalibration. To address it, we design two methods to improve calibration and performance in such scenarios. Motivated by the fact that predicted probability distributions of classes are highly related to the numbers of class instances, we propose label-aware smoothing to deal with different degrees of over-confidence for classes and improve classifier learning. For dataset bias between these two stages due to different samplers, we further propose shifted batch normalization in the decoupling framework. Our proposed methods set new records on multiple popular long-tailed recognition benchmark datasets, including CIFAR-10-LT, CIFAR-100-LT, ImageNet-LT, Places-LT, and iNaturalist 2018. Code will be available at https://github.com/Jia-Research-Lab/MiSLAS.

PDF Abstract CVPR 2021 PDF CVPR 2021 Abstract

Code

Add Remove Mark official

Jia-Research-Lab/MiSLAS official

140

dvlab-research/MiSLAS

140

dvlab-research/rescom

dvlab-research/imbalanced-learning

simonustc/imbalance_ppl_cri

Tasks

Add Remove

Long-tail Learning

Representation Learning

Datasets

CIFAR-10

CIFAR-100

Places

iNaturalist ImageNet-LT Places-LT

Results from the Paper

Edit

Ranked #16 on Long-tail Learning on CIFAR-10-LT (ρ=100)

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Long-tail Learning	CIFAR-100-LT (ρ=10)	MiSLAS	Error Rate	36.8	# 18	Compare
Long-tail Learning	CIFAR-100-LT (ρ=100)	MiSLAS	Error Rate	53	# 37	Compare
Long-tail Learning	CIFAR-100-LT (ρ=50)	MiSLAS	Error Rate	47.7	# 19	Compare
Long-tail Learning	CIFAR-10-LT (ρ=10)	MiSLAS	Error Rate	10	# 18	Compare
Long-tail Learning	CIFAR-10-LT (ρ=100)	MiSLAS	Error Rate	17.9	# 16	Compare
Long-tail Learning	ImageNet-LT	MiSLAS	Top-1 Accuracy	52.7	# 44	Compare
Long-tail Learning	iNaturalist 2018	MiSLAS	Top-1 Accuracy	71.6%	# 25	Compare

Methods

Add Remove

Batch Normalization

Edit Social Preview

Improving Calibration for Long-Tailed Recognition

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove