Application of Adam-BP neural network in leveling fitting

2019 2nd International Conference of Green Buildings and Environmental Management

生态环境科学

Dong, Zhouyang^1^2 ; Xu, Weiming^2 ; Xu, Jian^2 ; Zhuang, Hao^2

Dept. of Hydrography and Cartography, Dalian Naval Academy, Dalian

116000, China^1

Unit 32023, Dalian

116000, China^2

关键词: BP (back propagation) neural network;    BP neural networks;    Convergence rates;    Improved BP neural network;    Linear units;    Moment estimation;    Network activations;    Network optimization;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/310/2/022036/pdf
DOI  :  10.1088/1755-1315/310/2/022036

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

According to the accuracy of BP (Back Propagation) neural network in leveling fitting, an improved BP neural network model is proposed. In order to overcome the defect of the low convergence rate and local minimal in leveling fitting. The BP neural network is improved by using Relu (Rectified Linear Units) function as network activation function and Adam (Adaptive Moment Estimation) algorithm as network optimization function. The experimental results show that the improved BP neural network can effectively improve the accuracy.