PSigmoid: Improving squeeze-and-excitation block with parametric sigmoid

doi:10.1007/s10489-021-02247-z

CASIA OpenIR > 智能制造技术与系统研究中心 > 多维数据分析（彭思龙）-技术团队

	PSigmoid: Improving squeeze-and-excitation block with parametric sigmoid
	Ying, Yao 1; Zhang, Nengbo 2; Shan, Peng 1; Miao, Ligang 3; Sun, Peng 4; Peng, Silong5
发表期刊	APPLIED INTELLIGENCE
ISSN	0924-669X
	2021-03-09
页码	13
通讯作者	Shan, Peng(6094079@qq.com)
摘要	Squeeze-and-Excitation (SE) Networks won the last ImageNet Large-Scale Visual Recognition Challenge (ILSVRC) classification competition and is very popular in today's vision community. The SE block is the core of Squeeze-and-Excitation Network (SENet), which adaptively recalibrates channel-wise features and suppresses less useful ones. Since SE blocks can be directly used in existing models and effectively improve performance, SE blocks are widely used in a variety of tasks. In this paper, we propose a novel Parametric Sigmoid (PSigmoid) to enhance the SE block. We named the new module PSigmoid SE (PSE) block. The PSE block can not only suppress features in a channel-wise manner, but also enhance features. We evaluate the performance of our method on four common datasets including CIFAR-10, CIFAR-100, SVHN and Tiny ImageNet. Experimental results show the effectiveness of our method. In addition, we compare the differences between the PSE block and the SE block through a detailed analysis of the configuration. Finally, we use a combination of PSE block and SE block to obtain better performance.
关键词	Activation function Sigmoid Parametric sigmoid Squeeze-and-excitation Convolutional neural network
DOI	10.1007/s10489-021-02247-z
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61601104] ; Fundamental Research Funds for the Central Universities[N2023021]
项目资助者	National Natural Science Foundation of China ; Fundamental Research Funds for the Central Universities
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000626811800001
出版者	SPRINGER
引用统计	被引频次：7[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/44132
专题	智能制造技术与系统研究中心_多维数据分析（彭思龙）-技术团队
通讯作者	Shan, Peng
作者单位	1.Northeastern Univ, Coll Informat Sci & Engn, Shenyang 110819, Peoples R China 2.Chinese Acad Sci, Shenzhen Inst Adv Technol, Shenzhen 518055, Peoples R China 3.Northeastern Univ, Sch Comp & Commun Engn, Shenyang 110819, Peoples R China 4.Criminal Invest Police Univ China, Audio Visual & Image Technol Dept, Shenyang 110854, Peoples R China 5.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Ying, Yao,Zhang, Nengbo,Shan, Peng,et al. PSigmoid: Improving squeeze-and-excitation block with parametric sigmoid[J]. APPLIED INTELLIGENCE,2021:13.
APA	Ying, Yao,Zhang, Nengbo,Shan, Peng,Miao, Ligang,Sun, Peng,&Peng, Silong.(2021).PSigmoid: Improving squeeze-and-excitation block with parametric sigmoid.APPLIED INTELLIGENCE,13.
MLA	Ying, Yao,et al."PSigmoid: Improving squeeze-and-excitation block with parametric sigmoid".APPLIED INTELLIGENCE (2021):13.