Essential Building Blocks of Convolutional Neural Network for Deep Learning

• Author(s): Olasunkanmi Felix Oyadokun ; Danjuma Shadrach Sunday ; Haruna Bege ; Kolawole Samuel F
• Paper ID: 1703854
• Page: 89-94
• Published Date: 20-10-2022
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 6 Issue 4 October-2022

Abstract

Computations for Deep Learning (DL) are designed to mimic the functionality of the neurons found in the brains of mammals. To achieve this, DL utilizes different algorithms to learn patterns in number fed as input to the system. One among the algorithms it uses to achieve this is the Convolutional Neural Network (CNN). CNN is best used to manipulate images in order to enable machines learn the patterns in them. In this article, kernel or filter, stride, padding, pooling and flattening shall be considered as fundamental building blocks of CNN for DL.

Keywords

CNN, DL, Neural network, kernel, stride, padding, pooling, flattening

Citations

IRE Journals:
Olasunkanmi Felix Oyadokun , Danjuma Shadrach Sunday , Haruna Bege , Kolawole Samuel F "Essential Building Blocks of Convolutional Neural Network for Deep Learning" Iconic Research And Engineering Journals Volume 6 Issue 4 2022 Page 89-94

IEEE:
Olasunkanmi Felix Oyadokun , Danjuma Shadrach Sunday , Haruna Bege , Kolawole Samuel F "Essential Building Blocks of Convolutional Neural Network for Deep Learning" Iconic Research And Engineering Journals, 6(4)