卷积网络深度学习算法与实例  

Convolutional Neural Network Algorithm and Case

作　　者：陈旭 张军[1] 陈文伟[1] 李硕豪[1] Chen Xu;Zhang Jun;Chen Wen-wei;Li Shuo-hao(College of Information System and Management, National University of Defense Technology, Changsha 410073, China)

机构地区：[1]国防科技大学信息系统与管理学院,湖南长沙410073

出　　处：《广东工业大学学报》2017年第6期20-26,共7页Journal of Guangdong University of Technology

基　　金：国家自然科学基金资助项目(6117019)

摘　　要：卷积神经网络(CNN)和反向传播神经网络结合组成的深度学习模型,是目前应用比较广泛的深度学习模型,但是具体算法有多种.本文针对以CNN为主体的深度学习算法的实现技术,包括卷积核设计、池化的作用、激活函数的选取和训练过程等问题进行详细讨论,并结合实例进行说明,便于对卷积网络深度学习算法的掌握,使其能得到更广泛的应用.最后总结并展望卷积神经网络未来的研究方向.Convolutional neural network(CNN)is a deep learning model with strong expression and classification ability,and is currently widely used,but there are a variety of specific algorithms.In this paper,the realization of deep learning algorithm based on convolutional neural network CNN,including the function of convolution kernel,the role of pooling,the selection of activation function and the training process are discussed.And one example is explained,which facilitates the mastery of CNN.Finally,the future research direction of convolution neural network is summarized and forecasted.

关 键 词：卷积神经网络 反向传播 深度学习 卷积层 池化层 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]