arrow
返回

Trainable activation function with differentiable negative side and adaptable rectified point

delete2020-10-15
delete10
PRE
AI
P
Pratama Kevin
D
Dae-Ki Kang *
DOI:10.1007/s10489-020-01885-zdelete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Activation function is one of the keys for Artificial Neural Network in learning complex mapping function. A recently proposed activation function called Scaled Exponential Linear Unit (SELU) has a unique characteristic in the ability to automatically normalize its output toward predefined mean and variance. In this paper, we introduce Parametric Scaled Exponential Linear Unit (PSELU), a modification of SELU where the parameters are adaptively learned during the training phase via backpropagation algorithm. We then add further modifications with the aim of having a stronger gradient in the negative part and produce more negative output from our proposed method. Our proposed method is evaluated using various artificial neural network models on diverse image classification tasks. In addition, we also measure the performance of our proposed method in comparison with some other popular activation functions. The evaluation results in this paper present some empirical proof of improvement in term of network generalization performance. Moreover, our experiments strengthen our initial hypothesis regarding the advantages obtained from using our proposed method. Finally, the behavior of the trained parameters along with the training phase dynamic is observed in order to better understand the increased performance of our proposed method
Keyword:
Activation functions
Self-normalizing neural networks
Parameterized learning
Rectified point
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Applied Intelligence 封面图
Applied Intelligence
IF:
3.5
论文数:
7.6K
被引数:
1.7W

机构

D
Dongseo University
学者数:
355
论文数: 389
被引数: 228
引用论文

引用论文

Single image super-resolution using a polymorphic parallel CNN
err2018-08-25
err37
errOAAI
errZeng, Kai; Ding, Shifei; Jia, Weikuan
err分享
err收藏
err分享
err收藏
A novel softplus linear unit for deep convolutional neural networks
err2017-09-01
err46
PREAI
errZhao, Huizhen; Liu, Fuxian; Li, Longyue; Luo, Chang
err分享
err收藏
Image super-resolution with densely connected convolutional networks
err2018-07-12
err25
PREAI
errKuang, Ping; Ma, Tingsong; Chen, Ziwei; Li, Fan
err分享
err收藏
学者 查看更多内容