Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network

Artificial neural network (ANN) are widely used as an engineering approach to mimic the human brain activities. It has applied in different aspects such as pattern recognition, alphabet or digit classification, handwriting recognition, speech recognition, fingerprint identification, data mining, rob...

Full description

Bibliographic Details
Main Author: Kang, Miew How
Format: Thesis
Language:English
Published: 2016
Subjects:
Online Access:http://eprints.usm.my/41299/
http://eprints.usm.my/41299/1/KANG_MIEW_HOW_24_Pages.pdf
_version_ 1848879252714291200
author Kang, Miew How
author_facet Kang, Miew How
author_sort Kang, Miew How
building USM Institutional Repository
collection Online Access
description Artificial neural network (ANN) are widely used as an engineering approach to mimic the human brain activities. It has applied in different aspects such as pattern recognition, alphabet or digit classification, handwriting recognition, speech recognition, fingerprint identification, data mining, robots and etc. Back-propagation is the most common artificial neural training algorithm, however it is suffering with the slow convergence rate issue. A study to improve the slow convergence rate of a neural network without sacrificing the accuracy of the network are carried out. In this research, a hand-written character recognition model are implemented in C++ programming with ability to classify digits 0, 1, 2, and 3. This model are built up with 64-40-4 neural network where input data are 8 x 8 dimension image and output are classified to 4 digits which are 0, 1, 2 and 3. Two modified algorithms are proposed in this research, which are mixture of the momentum algorithm with different learning rate algorithms. First proposed algorithm is the combination of momentum algorithm with adaptive learning rate (ALR) algorithm, and second proposed algorithm is the combination of momentum algorithm with automatic learning rate selection (ALRS) algorithm. Convergence rate are showed 18% improvement in the algorithm mixture with ALR algorithm, however there is no significant improvement of timing for algorithm mixture with ALRS algorithm.
first_indexed 2025-11-15T17:44:17Z
format Thesis
id usm-41299
institution Universiti Sains Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T17:44:17Z
publishDate 2016
recordtype eprints
repository_type Digital Repository
spelling usm-412992018-08-13T06:54:12Z http://eprints.usm.my/41299/ Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network Kang, Miew How TK7800-8360 Electronics Artificial neural network (ANN) are widely used as an engineering approach to mimic the human brain activities. It has applied in different aspects such as pattern recognition, alphabet or digit classification, handwriting recognition, speech recognition, fingerprint identification, data mining, robots and etc. Back-propagation is the most common artificial neural training algorithm, however it is suffering with the slow convergence rate issue. A study to improve the slow convergence rate of a neural network without sacrificing the accuracy of the network are carried out. In this research, a hand-written character recognition model are implemented in C++ programming with ability to classify digits 0, 1, 2, and 3. This model are built up with 64-40-4 neural network where input data are 8 x 8 dimension image and output are classified to 4 digits which are 0, 1, 2 and 3. Two modified algorithms are proposed in this research, which are mixture of the momentum algorithm with different learning rate algorithms. First proposed algorithm is the combination of momentum algorithm with adaptive learning rate (ALR) algorithm, and second proposed algorithm is the combination of momentum algorithm with automatic learning rate selection (ALRS) algorithm. Convergence rate are showed 18% improvement in the algorithm mixture with ALR algorithm, however there is no significant improvement of timing for algorithm mixture with ALRS algorithm. 2016 Thesis NonPeerReviewed application/pdf en http://eprints.usm.my/41299/1/KANG_MIEW_HOW_24_Pages.pdf Kang, Miew How (2016) Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network. Masters thesis, Universiti Sains Malaysia.
spellingShingle TK7800-8360 Electronics
Kang, Miew How
Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network
title Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network
title_full Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network
title_fullStr Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network
title_full_unstemmed Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network
title_short Study Of Modified Training Algorithm For Optimized Convergence Speed Of Neural Network
title_sort study of modified training algorithm for optimized convergence speed of neural network
topic TK7800-8360 Electronics
url http://eprints.usm.my/41299/
http://eprints.usm.my/41299/1/KANG_MIEW_HOW_24_Pages.pdf