Curricular-balanced long-tailed learning

doi:10.1016/j.neucom.2023.127121

	Curricular-balanced long-tailed learning
	Xiang, Xiang 1; Zhang, Zihan 1; Chen, Xilin 2
	2024-02-28
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
卷号	571 页码:13
摘要	The real-world data distribution is essentially long-tailed, which poses a significant challenge to the deep model. Classification models minimizing cross-entropy loss struggle to classify the tail classes, although cross-entropy training is successful on balanced data. We reveal that minimizing cross-entropy loss under long-tailed distribution leads to the Tail Collapse phenomenon, which fundamentally limits the performance of neural networks. To correct the optimization behavior of cross-entropy training, we propose a new Curricular Balanced Loss (CurB Loss) to alleviate the imbalance. The CurB loss has two factors: the re-weighting factor and the curriculum learning factor. We design the re-weighting factor based on the margin-based training that can theoretically reach the optimums of networks. Then, we incorporate the idea of Curriculum Learning into the re-weighting loss in an adaptive manner. We design the curriculum learning factor to make the model gradually emphasize the hard classes. The empirical results demonstrate the complementary of the two factors. Our method outperforms previous state-of-the-art methods by 0.9%, 2.7%, 1.2% on CIFAR10-LT, CIFAR-100-LT and ImageNet-LT, demonstrating the effectiveness of CurB Loss for long-tailed visual recognition.
关键词	Long-tailed learning Re-weighting loss Curriculum learning
DOI	10.1016/j.neucom.2023.127121
收录类别	SCI
语种	英语
资助项目	Natural Science Fund of Hubei Province[2022CFB823] ; HUST Independent Innovation Research Fund[2021XXJS096] ; Alibaba Innovation Research Program[CRAQ7WHZ11220001-20978282]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:001148963100001
出版者	ELSEVIER
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38400
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Xiang, Xiang
作者单位	1.Huazhong Univ Sci & Technol HUST, Sch Artificial Intelligence & Automat, Natl Key Lab Multispectral Informat Intelligent Pr, Wuhan, Peoples R China 2.Chinese Acad Sci, Key Lab Intelligent Informat Proc, Inst Comp Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Xiang, Xiang,Zhang, Zihan,Chen, Xilin. Curricular-balanced long-tailed learning[J]. NEUROCOMPUTING,2024,571:13.
APA	Xiang, Xiang,Zhang, Zihan,&Chen, Xilin.(2024).Curricular-balanced long-tailed learning.NEUROCOMPUTING,571,13.
MLA	Xiang, Xiang,et al."Curricular-balanced long-tailed learning".NEUROCOMPUTING 571(2024):13.