中国计量大学学报2017，Vol.28Issue(4)：485-491,7.DOI:10.3969/j.issn.2096-2835.2017.04.013
矩阵输入的多层前向神经网络学习算法
Learning algorithm for multilayer feed-forward neural networks with matrix inputs
摘要
Abstract
The learning ability of single hidden layer feed-forward neural networks was limited .In particular, as a classifier ,for single hidden layer feed-forward neural networks it was difficult to learn and handle the complex information and the details of different pictures.Driven by the mind of the deep neural networks theory ,the single hidden layer neural networks with matrix inputs were extended to the case of multi hidden layers.In addition,the extended networks were trained by the traditional back propagation algorithm ,and a corresponding learning algorithm was obtained .The experimental results on several databases show that the proposed algorithm possesses good performance compared with some existing methods.关键词
神经网络/图像分类/深度学习Key words
neural networks/image classification/deep learning分类
信息技术与安全科学引用本文复制引用
黄旭进,曹飞龙..矩阵输入的多层前向神经网络学习算法[J].中国计量大学学报,2017,28(4):485-491,7.基金项目
国家自然科学基金资助项目(No.61672477). （No.61672477）
