eprintid: 15447
rev_number: 2
eprint_status: archive
userid: 1
dir: disk0/00/01/54/47
datestamp: 2023-11-10 03:30:04
lastmod: 2023-11-10 03:30:04
status_changed: 2023-11-10 01:59:30
type: conference_item
metadata_visibility: show
creators_name: Al-Saggaf, U.M.
creators_name: Botalb, A.
creators_name: Moinuddin, M.
creators_name: Alfakeh, S.A.
creators_name: Ali, S.S.A.
creators_name: Boon, T.T.
title: Either crop or pad the input volume: What is beneficial for Convolutional Neural Network?
ispublished: pub
keywords: Convolution; Deep learning, Convolutional kernel; Convolutional neural network; Cropping; Hyper-parameter; Kernel size; Number of layers; OR pads; Padding; Pooling; Window Size, Convolutional neural networks
note: cited By 1; Conference of 8th International Conference on Intelligent and Advanced Systems, ICIAS 2021 ; Conference Date: 13 July 2021 Through 15 July 2021; Conference Code:175661
abstract: Convolutional Neural Network (CNN) is the most popular method of deep learning in the machine learning field. Training a CNN has always been a demanding task compared to other machine learning paradigms, and this is due to its big space of hyper-parameters such as convolutional kernel size, number of strides, number of layers, pooling window size, etc. What makes the CNN's huge hyper-parameters space optimization harder is that there is no universal robust theory supporting it, and any work flow proposed so far in literature is based on heuristics that are just rules of thumb and only depend on the dataset and problem at hand. In this work, it is empirically illustrated that the performance of a CNN is not linked only with the choice of the right hyper-parameters, but also linked to how some of the CNN operations are implemented. More specifically, the CNN performance is contrasted for two different implementations: cropping and padding the input volume. The results state that padding the input volume achieves higher accuracy and takes less time in training compared with cropping method. © 2021 IEEE.
date: 2021
publisher: Institute of Electrical and Electronics Engineers Inc.
official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85124170608&doi=10.1109%2fICIAS49414.2021.9642661&partnerID=40&md5=fe97d0334bee476e8e7e0756fc5b0943
id_number: 10.1109/ICIAS49414.2021.9642661
full_text_status: none
publication: International Conference on Intelligent and Advanced Systems: Enhance the Present for a Sustainable Future, ICIAS 2021
refereed: TRUE
isbn: 9781728176666
citation:   Al-Saggaf, U.M. and Botalb, A. and Moinuddin, M. and Alfakeh, S.A. and Ali, S.S.A. and Boon, T.T.  (2021) Either crop or pad the input volume: What is beneficial for Convolutional Neural Network?  In: UNSPECIFIED.