摘要
ResNet的瓶颈结构存在一定的冗余输出,面对类别较少的分类任务时,冗余信息占比会非常大,造成计算资源浪费。针对这个问题,该文利用残差结构和连接操作设计了一个新的升维结构(RC结构)用于改进瓶颈结构。RC结构可以降低瓶颈结构的资源消耗,并增强梯度传递。将RC与多种残差网络进行结合,并在多个数据集上进行图像分类实验。实验结果表明,基于RC的残差网络在面对类别较少的分类任务时,能有效提高网络的效率和精度。
Research shows that there is redundancy in the output feature maps of ResNet bottleneck structure.Such redundancy ensures a comprehensive understanding of the input data,but generates the redundant information which consumes additional computational resources.And the proportion of redundant information is very large when processing small category classification tasks.To solve this problem,a new dimension increasing structure is designed to improve the bottleneck structure by residual-like structure and concatenation operation.This structure is called residual concatenation(RC).The RC can not only reduce the amount of calculation and parameters of bottleneck structure,but also enhance the gradient transmission of back-propagation to improve the accuracy.In this work,the RC is combined with multiple residual networks and image classification experiments are performed on multiple datasets.The results show that the RC-based bottleneck structure can reduce the consumption and improve the accuracy of classification tasks while processing small category classification tasks.
作者
储岳中
汪佳庆
张学锋
刘恒
CHU Yue-zhong;WANG Jia-qing;ZHANG Xue-feng;LIU Heng(School of Computer Science and Technology,Anhui University of Technology,Maanshan Anhui,243032;Anhui Engineering Laboratory for Intelligent Applications and Security of Industrial Internet,Maanshan Anhui,243032)
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
2021年第2期243-248,共6页
Journal of University of Electronic Science and Technology of China
基金
国家自然科学基金(61971004)。
