Abstract: The paper proposes the dynamic modification of the activation function in a learning technique, more exactly backpropagation algorithm. The modification consists in changing slope of sigmoid function for activation function according to increase or decrease the error in an epoch of learning. The study was done using the Waikato Environment for Knowledge Analysis (WEKA) platform to complete adding this feature in Multilayer Perceptron class. This study aims the dynamic modification of activation function has changed to relative gradient error, also neural networks with hidden layers have not used for it.

Marina Adriana Mercioni, Alexandru Tiron and Stefan Holban, “Dynamic Modification of Activation Function using the Backpropagation Algorithm in the Artificial Neural Networks” International Journal of Advanced Computer Science and Applications(IJACSA), 10(4), 2019. http://dx.doi.org/10.14569/IJACSA.2019.0100406

@article{Mercioni2019,
title = {Dynamic Modification of Activation Function using the Backpropagation Algorithm in the Artificial Neural Networks},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2019.0100406},
url = {http://dx.doi.org/10.14569/IJACSA.2019.0100406},
year = {2019},
publisher = {The Science and Information Organization},
volume = {10},
number = {4},
author = {Marina Adriana Mercioni and Alexandru Tiron and Stefan Holban}
}