Cover Image

A comparative performance of different convolutional neural network activation functions on image classification

Activation functions are crucial in optimising Convolutional Neural Networks (CNNs) for image classification. While CNNs excel at capturingspatial hierarchies in images, the activation functions substantially impact their effectiveness. Traditional functions, such as ReLU and Sigmoid, have drawbacks...

Full description

Saved in:
Bibliographic Details
Main Authors: Azhary, Muhammad Zulhazmi Rafiqi, Ismail, Amelia Ritahani
Format: Article
Language:English
Published: IIUM Press 2024
Subjects:
QA75 Electronic computers. Computer science
Online Access:http://irep.iium.edu.my/116734/7/116734_A%20comparative%20performance.pdf
http://irep.iium.edu.my/116734/
https://journals.iium.edu.my/kict/index.php/IJPCC/article/view/490/295
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.iium.irep.116734
record_format dspace
spelling my.iium.irep.1167342024-12-17T08:35:04Z http://irep.iium.edu.my/116734/ A comparative performance of different convolutional neural network activation functions on image classification Azhary, Muhammad Zulhazmi Rafiqi Ismail, Amelia Ritahani QA75 Electronic computers. Computer science Activation functions are crucial in optimising Convolutional Neural Networks (CNNs) for image classification. While CNNs excel at capturingspatial hierarchies in images, the activation functions substantially impact their effectiveness. Traditional functions, such as ReLU and Sigmoid, have drawbacks, including the "dying ReLU" problem and vanishing gradients, which can inhibit learning and efficacy. The study seeks to comprehensively analyse various activation functions across different CNN architectures to determine their impact on performance. The findings suggest that Swish and Leaky ReLU outperform other functions, with Swish particularly promising in complicated networks such as ResNet. This emphasises the relevance of activation function selection in improving CNN performance and implies that investigating alternative functions can lead to more accurate and efficient models for image classification tasks. IIUM Press 2024-07-30 Article PeerReviewed application/pdf en http://irep.iium.edu.my/116734/7/116734_A%20comparative%20performance.pdf Azhary, Muhammad Zulhazmi Rafiqi and Ismail, Amelia Ritahani (2024) A comparative performance of different convolutional neural network activation functions on image classification. International Journal on Perceptive and Cognitive Computing (IJPCC), 10 (2). pp. 118-122. E-ISSN 2462-229X https://journals.iium.edu.my/kict/index.php/IJPCC/article/view/490/295 10.31436/ijpcc.v10i2.490
institution Universiti Islam Antarabangsa Malaysia
building IIUM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider International Islamic University Malaysia
content_source IIUM Repository (IREP)
url_provider http://irep.iium.edu.my/
language English
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Azhary, Muhammad Zulhazmi Rafiqi
Ismail, Amelia Ritahani
A comparative performance of different convolutional neural network activation functions on image classification
description Activation functions are crucial in optimising Convolutional Neural Networks (CNNs) for image classification. While CNNs excel at capturingspatial hierarchies in images, the activation functions substantially impact their effectiveness. Traditional functions, such as ReLU and Sigmoid, have drawbacks, including the "dying ReLU" problem and vanishing gradients, which can inhibit learning and efficacy. The study seeks to comprehensively analyse various activation functions across different CNN architectures to determine their impact on performance. The findings suggest that Swish and Leaky ReLU outperform other functions, with Swish particularly promising in complicated networks such as ResNet. This emphasises the relevance of activation function selection in improving CNN performance and implies that investigating alternative functions can lead to more accurate and efficient models for image classification tasks.
format Article
author Azhary, Muhammad Zulhazmi Rafiqi
Ismail, Amelia Ritahani
author_facet Azhary, Muhammad Zulhazmi Rafiqi
Ismail, Amelia Ritahani
author_sort Azhary, Muhammad Zulhazmi Rafiqi
title A comparative performance of different convolutional neural network activation functions on image classification
title_short A comparative performance of different convolutional neural network activation functions on image classification
title_full A comparative performance of different convolutional neural network activation functions on image classification
title_fullStr A comparative performance of different convolutional neural network activation functions on image classification
title_full_unstemmed A comparative performance of different convolutional neural network activation functions on image classification
title_sort comparative performance of different convolutional neural network activation functions on image classification
publisher IIUM Press
publishDate 2024
url http://irep.iium.edu.my/116734/7/116734_A%20comparative%20performance.pdf
http://irep.iium.edu.my/116734/
https://journals.iium.edu.my/kict/index.php/IJPCC/article/view/490/295
_version_ 1818833733419008000
score 13.223943

Similar Items