一种改进激活函数的人工神经网络及其应用

基金项目: 国家自然科学基金资助项目(29875021)。

详细信息

An Artificial Neural Network with Improved Activation Function and Its Application

1 School of Electronic Information, Wuhan University, 129 Luoyu Road, Wuhan 430079, China;
2 College of Chemistry and Molecular Sciences, Wuhan University, Luojia Hill, Wuhan 430072, China

摘要: 提出了人工神经元的一种新颖的多参数可调激活函数,推导出相应的BP学习算法。在人工神经网络用于除草剂化合物活性预测的研究中,和传统BP算法的对比试验显示,本文的改进BP网络具有更快的收敛速度和更高的精度。
- 人工神经网络 /
- 激活函数 /
- BP算法
Abstract: In this paper, a novel variant activation (transform) sigmoid function with three parameters is proposed, and then the improved BP algorithm based on it is educed and discussed. The efficiency and advantage of the method is proved by the prediction results for the activity of herbicide based on the improved and traditional BP ANNs.
- artificial neural network /
- activation function /
- BP algorithm