The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
L'une des raisons de la lente convergence dans l'apprentissage par rétropropagation est la valeur décroissante de la dérivée des fonctions d'activation couramment utilisées à mesure que les nœuds s'approchent des valeurs extrêmes, à savoir 0 ou
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Joarder KAMRUZZAMAN, "Arctangent Activation Function to Accelerate Backpropagation Learning" in IEICE TRANSACTIONS on Fundamentals,
vol. E85-A, no. 10, pp. 2373-2376, October 2002, doi: .
Abstract: One of the reasons of slow convergence in Backpropagation learning is the diminishing value of the derivative of the commonly used activation functions as the nodes approach extreme values, namely, 0 or
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e85-a_10_2373/_p
Copier
@ARTICLE{e85-a_10_2373,
author={Joarder KAMRUZZAMAN, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Arctangent Activation Function to Accelerate Backpropagation Learning},
year={2002},
volume={E85-A},
number={10},
pages={2373-2376},
abstract={One of the reasons of slow convergence in Backpropagation learning is the diminishing value of the derivative of the commonly used activation functions as the nodes approach extreme values, namely, 0 or
keywords={},
doi={},
ISSN={},
month={October},}
Copier
TY - JOUR
TI - Arctangent Activation Function to Accelerate Backpropagation Learning
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2373
EP - 2376
AU - Joarder KAMRUZZAMAN
PY - 2002
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E85-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 2002
AB - One of the reasons of slow convergence in Backpropagation learning is the diminishing value of the derivative of the commonly used activation functions as the nodes approach extreme values, namely, 0 or
ER -