Dynamic Modification of Activation Function using the Backpropagation Algorithm in the Artificial Neural Networks

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.

Authors and Affiliations

Marina Adriana Mercioni, Alexandru Tiron, Stefan Holban

Keywords

Related Articles

Relationship Analysis on the Experience of Hospitalised Paediatric Cancer Patient in Malaysia using Text Analytics Approach

The purpose of this study is to analyse the keyword relationships of paediatric cancer patient’s experiences whilst being hospitalised during the treatment session. This study collects data through 40 days of observation...

A General Evaluation Framework for Text Based Conversational Agent

This paper details the development of a new evaluation framework for a text based Conversational Agent (CA). A CA is an intelligent system that handle spoken or/and text based conversations between machine and human. Gen...

Detection of Infected Leaves and Botanical Diseases using Curvelet Transform

The study of plants is known as botany and for any botanist it is a daily routine work to examine various plants in their research lab. This research efforts an image processing-based algorithm for extracting the region...

Towards Usability Guidelines for the Design of Effective Arabic Websites

The Arabic websites constitute 1% of the web content with more than 225 million viewers and 41% Internet penetration. However, there is a lack of design guidelines related to the selection and use of appropriate font typ...

 Coordinate Rotation Digital Computer Algorithm: Design and Architectures

 COordinate Rotation DIgital Computer (CORDIC) algorithm has potential for efficient and low-cost implementation of a large class of applications which include the generation of trigonometric, logarithmic and transc...

Download PDF file
  • EP ID EP550253
  • DOI 10.14569/IJACSA.2019.0100406
  • Views 114
  • Downloads 0

How To Cite

Marina Adriana Mercioni, Alexandru Tiron, Stefan Holban (2019). Dynamic Modification of Activation Function using the Backpropagation Algorithm in the Artificial Neural Networks. International Journal of Advanced Computer Science & Applications, 10(4), 51-56. https://europub.co.uk/articles/-A-550253