A Method to Solve overflow or underflow errors Resulting from Activation functions in Convolutional Neural Networks

In this paper, a method is proposed to correct overflow errors of resulting values that can occur when the cross-entropy loss function is used as a softmax and ReLU as an activation function of the convolutional neural network. Since conventional softmax and cross-entropy functions include exponenti...

Full description

Bibliographic Details

Published in:	International Information Institute (Tokyo). Information Vol. 20; no. 11; pp. 8151 - 8158
Main Authors:	Jeong, JunMo, Choi, SeJin, Kim, ChiYong
Format:	Journal Article
Language:	English
Published:	Koganei International Information Institute 01.11.2017
Subjects:	Activation Algorithms Artificial neural networks Classification Divergence Entropy Error correction Methods Neural networks Neurons
ISSN:	1343-4500, 1344-8994
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!